Background
Hip fractures are a common and major health problem among older people [
1]. The annual number of such fractures in Sweden is expected to almost double during the first half of this century [
2]. There is a well-established increased risk of death after hip fracture [
3,
4]. It has been shown that older people have a 5- to 8-fold increased risk of dying during the first 3 months after a hip fracture [
4]. Studies have been performed to optimize care of hip fracture patients and the consensus concerning preoperative management, time to surgery, operative management, surgical technique and postoperative care has led to recommendations concerning clinical care pathways and a multidisciplinary approach [
5‐
8]. Despite research into care improvements for hip fracture patients, for example in the fields of medical and surgical care or the use of multidisciplinary teams, it has been shown that fewer than half regain their previous level of function [
9] and the mortality rate has remained stable over the past 40 years [
10].
A variety of postoperative complications and times to follow-up are described. One study found that 33 % of the participants had at least one complication after an operation for hip-fracture which led to prolonged hospitalization. The most common postoperative complications were delirium and infection [
11]. Another study found heart failure and chest infections to be the most common postoperative complications [
12], while a more recent study showed that falls, fractures and pneumonia were the most common [
13].
A range of interventions to reduce the rate of in-hospital postoperative complications and mortality have been reported in the literature. One such study found that postoperative complications and 12-month mortality among community dwellers were reduced when a comprehensive multidisciplinary fast-track treatment and care program were put in place [
14]. We have shown earlier that a multidisciplinary, multi-factorial rehabilitation program reduced in-hospital complications [
15], including significant fewer in hospital falls but there was no difference in the number of falls during 1 year after discharge [
16,
17].
Death following hip fracture has been associated with several risk factors; older age and male sex, severe systemic disease, pre-fracture functional impairment, cognitive decline, coronary heart disease and the number of co-morbidities [
18‐
23]. The causes of excess death have been a subject of debate. One study suggests that the increased mortality is associated with postoperative complications [
24], others ascribe it to pre-fracture co-morbidities together with postoperative complications [
12,
19], or suggest that the co-morbidities are the underlying cause [
25]. Only a few studies describe the causes of death [
24,
26‐
29]. The most common causes of death in one autopsy study were chest infection, cardiac failure, myocardial infarction and pulmonary embolism [
29]. Among nursing-home residents with hip fracture the most common causes of death were infection, dementia and cardiac events [
26] and, in more recent studies, cardiac and infectious diseases [
27,
28].
In spite of earlier research, the poor outcome for people with hip fracture has not improved and mortality has not been reduced. Since neither the events leading to death, nor the causes of death or the patient’s outcome after discharge have been fully investigated, we decided to explore these factors in order to discover factors that might be adjusted in order to improve outcome.
Thus, the aim of this study was to describe the prevalence of co-morbidities, complications and causes of death and to investigate factors that could predict mortality in old people with femoral neck fracture.
Discussion
This study shows that both co-morbidities at baseline and complications during hospitalization are associated with mortality. Cancer, dependence in P-ADL, cardiovascular disease and dementia at baseline, and pulmonary emboli and cardiac failure during hospitalization were all independent predictors of mortality. Forty percent had died after 3 years despite the exclusion of those who were bedridden, had severe renal failure or pathological fractures. The most common primary causes of death were cardiovascular events, dementia, fractures, cancer and cerebrovascular events. The participants had several co-morbidities and suffered numerous complications such as infections, falls and fractures, cardiovascular events, delirium, and pressure ulcers.
When analysing factors associated with death we found that both comorbidities and post-operative complications were of significance, a finding which is also verified in a recent review using data from a National Trauma Data Bank [
35]. A study by Roche et al. [
12] found that age, male sex, cancer, chest infection, cardiac failure and stroke could predict mortality but they did not include dementia or functional measurements in their model. Dementia, however, is common among hip fracture patients and was thus included in the present study. Dementia is also found to be a risk factor for death and in another large cohort study by Petersen et al. [
19] age, cardiac complications and dementia were associated with mortality at 12 months but malignancies, cardiovascular disease and measurements of function were not included. In a large cohort study by Castronuovo et al. [
36] heart disease was a risk factor for 30-days mortality but not during follow-up though no complications were included in the study.
A reduced mental and medical status, and a poor physical ability at baseline were found among the deceased in the current study, similar to the results of a Norwegian study [
23], although those living in nursing homes and patients who did not pass a mental status test had been excluded. Such patients were not excluded in the present study. Among those who died, 56 % had three or more co-morbidities at baseline and 65 % had an ASA score of 3 or higher. The number of co-morbidities and poor pre-fracture status might be indicators of frailty. It has been shown that the (ASA) classification of medical co-morbidities is strongly associated with medical problems in the perioperative period [
37] and an earlier Swedish study shows that a high ASA score is a factor associated with mortality [
21]. A cardiovascular disease is a strong predictor of post-operative cardiac failure, according to Roche et al. [
12]. All participants in the present study who died during hospitalization due to a cardiovascular events had a pre-fracture cardiovascular disease.
Early death within 30 days after admission to hospital due to a hip fracture has recently been described in a study that examines post-mortem reports, where respiratory infections and cardiovascular disease were the main causes of death [
28]. These results are in line with earlier studies [
24,
29] and our study shows a similar result as 46 % of early deaths were due to a cardiovascular event. During follow-up cardiovascular events, dementia and cancer were the most common causes of death in the present study, which is partly consistent with earlier studies [
26,
27,
38]. The difference in the prevalence of infection as a cause might be due to the manner in which assessment of the causes of death was determined in the present study, as described above in the method section. There might also be an under diagnosis of dementia among many old people [
39]. The participants in the present study were cognitively tested during hospitalization and at 4, 12 and 36 months.
Our study confirms that complications among people suffering from a hip fracture are numerous, both early post-operatively and during follow-up. In-hospital complications among hip fracture patients, such as urinary tract infections, delirium, decubital ulcers and falls, can be successfully prevented and treated [
14,
15] but as far as we know few intervention studies have succeeded in preventing cardiac complications [
40‐
42].
Since we can now identify the most vulnerable patients, the focus in further research should be on prevention of infections and heart diseases in early postoperative care since they might be avoidable. These findings are in line with Petersen et al. [
19] who concluded that cardiac complications constituted an important risk factor that might be modified and with Roche et al. [
12] who emphasized the need for medical assessment among those with heart failure and chest infection.
In addition, prevention should also focus on the numerous complications that occur after discharge from hospital. General complications were associated with loss of function in a recent study by Hansson et al. [
13] Improved rehabilitation after stroke, including treatment of underlying comorbidities as well as secondary prevention, has increased the survival rate after stroke over the last few decades [
43] while mortality after femoral neck fracture has remained constant. Treatment of risk factors for stroke and myocardial infarction as well as secondary prevention are currently well established in routine care. The 56 new fractures that occurred in the present sample indicate that fracture prevention also needs to become a part of routine care aimed at reducing fracture rates and mortality.
Since hip fracture is, in many cases, an event that signals a systemic decline in the person’s health, it is crucial to preserve and improve the clinical care pathways to ensure optimal recovery and survival for these patients. The present study will, hopefully, contribute to the knowledge available concerning the causes of death and highlight the potentially modifiable/preventable complications that we need to focus on in the future.
The strength of our study is that we have systematically analysed all complications and the secondary causes of death among people with hip fracture, which no other study has reported to our knowledge. Differences in the present results compared to other studies might be due to the definitions applied to complications and co-morbidities and to the choice of exclusion criteria. In the future a more standardized description of the samples would facilitate comparison. In the present study we tried to obtain information about all events that the participants experienced and that led to death. An underlying cause of death is defined as “the disease or injury which initiated the train of morbid events leading directly to death, or the circumstances of the accident or violence which produced the fatal injury”, in accordance with the rules of the ICD. Although the intention of the ICD is to provide a standard means of recording underlying causes of death, comparison of cause-of-death data over time and across countries should be undertaken with caution. The rules for selecting the underlying cause of death have been re-evaluated and sometimes changed. Incorrect or incomplete death certificates, misinterpretation of ICD rules for selection of the underlying cause, and variations in the use of coding categories for unknown and ill-defined causes might all occur, according to the WHO.
There are some limitations to the present study. The sample was relatively small and the participants were only assessed three times over 3 years after discharge and there are certainly complications that were missed, despite the thorough reviews of the participants’ medical records. As people with vertebral fractures and rib fractures do not always seek medical care such fractures are poorly documented. The number of vertebral and rib fractures has probably been underestimated in our study since x-rays were not routinely taken during follow-up. Suffering pulmonary emboli is a serious condition and the HR for pulmonary emboli in the multivariate analyse should be interpreted with precaution as only one person in this sample had an emboli and died soon after the fracture.
Acknowledgements
We thank Maria Persson and Eva Elinge for her help with data collection.