Approach to patients with hip fracture and concurrent stroke

Abstract

We report a case of a 70-year-old man who presented with concomitant hip fracture and stroke. Our patient underwent surgical correction of a hip fracture despite the increased perioperative and postoperative risks associated with an acute stroke. He achieved good functional outcome after surgery and subsequent rehabilitation. There are no clear guidelines on the factors to determine whether a patient with concomitant stroke and hip fracture is a good candidate for surgical hip repair. Furthermore, there is also no consensus on the appropriate timing of surgical repair for such patients. We postulate that factors such as functional status, comorbidities, type and severity of stroke will affect the decision to proceed with surgical repair, and that there is a benefit in advocating for surgery in appropriate patients by a multidisciplinary orthogeriatric care team.

Background

Both hip fractures and stroke affect mainly the older age group. There were an estimated 13.6 million incident cases of stroke occurring worldwide in 2016.¹ Hip fractures are estimated to account for 0.1% of the global burden of disease, with an estimated 1.26 million cases in 1990, which is expected to double to 2.6 million by 2025.² Older patients discharged from an acute hospital with a stroke or hip fracture will usually require postacute rehabilitation either at home or in a rehabilitation facility. It is estimated that postacute care for conditions such as stroke and hip fracture may account for up to 15% of healthcare expenditure.³ Both conditions result in significant morbidity and mortality. While stroke mortality has declined in the past century due to improvements in care models, stroke remains the second leading cause of death worldwide after ischaemic heart disease.¹ Even stroke survivors with no early complications of stroke still have a 5.8% risk of death and a 1.3% risk of admission to a long-term care facility at 1 year.⁴ The 1-year mortality from a hip fracture is estimated to be as high as 30%, and up to 50% of the surviving patients remain institutionalised at 1 year.⁵

In light of the increasing healthcare burden of both hip fracture and stroke around the world, these two conditions have garnered increasing interest in the promotion of value-based healthcare through the introduction of bundled payment schemes aimed at improving cost-effectiveness and streamlining transitions of care from hospital to home.^{6 7} While this has promoted the development of guidelines and pathways for the management of patients who had a stroke and patients with hip fracture independently, there is still significant controversy in the management of patients presenting with concomitant stroke and hip fracture. Stroke is a risk factor for subsequent hip fracture particularly on the hemiplegic limb,⁸ postulated to be due to disuse osteoporosis in the hemiplegic limb and an increased risk of falls from weakness,⁹ impaired balance and impaired cognition.¹⁰ Most data suggest that approximately 7.3%–15.3% of all hip fractures occur in patients who had a stroke.^11–15 Hip fractures are also noted to be an independent risk factor for subsequent stroke, and patients with hip fracture have an approximately 1.5-fold increased risk of developing a stroke compared with a fracture-free cohort.¹⁶ In a Danish nationwide cohort study by Pedersen et al, the incidence of stroke at 30 days was 2.16% in patients with hip fracture compared with 0.21% in the general population.¹⁷

The challenges of managing concomitant stroke and hip fracture are compounded when the two pathologies present in close proximity, often encountered when the hip fracture is sustained as a result of a fall due to a stroke or when the stress of an osteoporotic hip fracture precipitates a stroke.

The concomitant presentation of a stroke with a hip fracture has various implications. First, the risk of surgical intervention to repair the hip is increased, with an increased risk of a recurrent stroke, myocardial infarction and even death.¹⁸ Second, patients with concomitant stroke and hip fracture are less likely to regain the expected functionality compared with patients without stroke. The overall frailty and functional status of the patient before admission becomes an important determinant in the decision to proceed with surgical intervention.

Case presentation

We report the case of a 70-year-old man with a medical history of diabetes mellitus, hypertension, hyperlipidaemia, and bilateral cataracts previously treated with intraocular lenses.

He was cognitively intact, independent in activities of daily living, but requiring a walking stick for ambulation due to osteoarthritis in both knees.

He presented with a fall while ambulating to a coffee shop near his home. Prior to the fall, he experienced a new-onset left-sided weakness. He did not complain of any slurring of speech, facial droop or blurring of vision, nor did he experience any dizziness, loss of consciousness, chest pain, palpitations or shortness of breath. He fell backwards and landed onto his buttocks. He was unable to get back up after the fall because of right-sided hip pain. He also had been suffering from a cough and fever during the week before his fall; cough was productive of yellow sputum with mild rhinorrhoea. He had not taken his long-term medications for the previous 3 weeks because he had run out. On admission, he was noted to be febrile with a temperature of 38.3°C and hypertensive with a blood pressure of 209/111 mm Hg. Other observations were otherwise unremarkable. A chest radiograph showed air space opacities in the right middle zone, suggestive of an infection. A hip radiograph showed a right intertrochanteric fracture (figure 1). An MRI of the brain done on admission showed a small acute right corona radiata infarct extending into the right internal capsule (figure 1). Neurological examination revealed normal motor power in the upper limbs (5/5). The power in the right lower limb was difficult to assess in view of pain from his concurrent hip fracture, and power over the left lower limb was diminished (3/5). He was subsequently admitted to the acute stroke unit and treated for an acute stroke as well as rhinovirus-positive community-acquired pneumonia.

Figure 1.

(A) T2-weighted MRI diffusion-weighted imaging done on admission showing a small focus of restricted diffusion in the right corona radiata. (B) Radiograph of right hip showing a minimally displaced right intertrochanteric fracture. (C) Radiograph of right hip after right proximal femoral nailing.

He was referred to orthopaedic surgery to explore surgical correction of the intertrochanteric hip fracture. The attending neurologist highlighted concerns regarding a high risk of stroke in the first 30 days postoperatively. Anaesthesia opinion concurred with that of neurology—that the patient is at a high risk of recurrent stroke in the perioperative period particularly if antiplatelet medications are discontinued for surgery, as well as at an increased risk of postoperative complications in view of his recent chest infection. Despite the high risks, the patient was initially keen for surgery as he was determined to regain his mobility. He was self-caring and independent at baseline, and he had no family available to care for him should he not be able to ambulate. Orthopaedics explained to the patient that the hip fracture could potentially heal without surgery. Furthermore, the risk of surgery was high in view of his concurrent acute stroke, and that risks of anaesthesia carried a greater than 10% risk of morbidity and mortality. On further consideration, our patient subsequently declined surgery due to the high risks explained and was transferred to a geriatric medicine subacute ward for rehabilitation.

While under the care of geriatric medicine, he continued to have non-remitting pain despite high doses of analgesia. Surgery was re-explored with orthopaedics. Despite the high risks, he decided to proceed with surgical fixation in view of persistent pain. He underwent a right proximal femoral nailing approximately 1 month after he sustained his fracture. Postoperatively, his stay was complicated by tachycardia secondary to hypovolaemia that resolved with intravenous hydration, and another episode of pneumonia was treated with antibiotics. He was subsequently transferred stably to an inpatient rehabilitation facility for further slow stream rehabilitation.

He was discharged from the inpatient rehabilitation facility after 2 months of rehabilitation. He had regained motor power over the lower limbs and was able to ambulate independently with a frame for 20 m. He had been provided with a motorised scooter to help him mobilise in the community. At his 3-month follow-up appointment, he had been coping well in the community independently. He reflected on his choice to undergo surgical correction of his hip fracture. He remains cognisant of the risks he undertook when he decided to proceed with the surgery. He considered it a ‘gamble’, but he was reassured by the fact that he had a neighbour who also sustained a hip fracture who had good functional outcomes after surgical correction.

Global health problem list

Predetermined pathways and value-based care in the management of stroke and hip fracture may lead to the underserving of complex high-needs patients such as those presenting concurrently with both conditions, particularly when no single discipline is identified to advocate in the interest of such patients.

There is a lack of consensus and evidence for a system to risk-stratify patients with concomitant hip fracture and stroke when deciding between operative and non-operative management, and there is a paucity of guidelines to help determine the ideal time frame in which to carry out surgery in such patients.

Global health problem analysis

The concomitant presentation of hip fracture and stroke leads to significant dilemmas in planning treatment, often resulting in different views from various specialties, most prominently neurology, anaesthesia, orthopaedics and geriatric medicine. There are primarily two difficult decisions that need to be made: first, the decision of whether to proceed with surgery in the context of an acute stroke; second, the decision of deciding the ideal time frame within which surgery should take place.

Approximately 90% of hip fractures are treated surgically,¹⁹ as non-operative management is associated with prolonged length of stay and higher incidence of morbidity and mortality.^{20 21} It should be noted that non-operative management may result in lower costs that are typically associated with direct surgical intervention and surgical complications and may be a reasonable choice in patients who are not medically fit for surgery.²² There is evidence to suggest that should non-operative management be opted for in patients who are not medically fit for surgery, there is benefit to early mobilisation to minimise complications of prolonged bed rest and reduce the risk of mortality.¹⁹ However, it should be noted that early mobilisation of unoperated patients also results in a more painful and prolonged recovery.¹⁹ A consideration on whether to proceed with surgery may be dependent on whether the patient can be mobilised without excessive pain even without surgery. There is limited evidence comparing the outcomes of non-operative versus operative treatment for patients who had a stroke with hip fractures. A study conducted in Chengdu on 238 patients who had a poststroke hemiplegia with poststroke trochanteric fractures showed that gamma nail internal fixation leads to improvements in time to ambulation, attenuation of complications of immobility and improved quality of life compared with non-operative management.²³

For patients who present with concomitant stroke and hip fracture, it will be important to identify a system to risk-stratify patients when deciding between operative and non-operative management. Operative management of a hip fracture primarily has the following objectives: to offer pain relief and to improve the likelihood of functional recovery. Factors to consider in deciding between operative and non-operative management include premorbid function, other comorbidities, type of stroke, overall prognosis and social support. The premorbid function of the patient is important in determining whether there is benefit in surgical intervention. Patients who are immobile or bedbound before surgery are not likely to show functional improvement after surgery.²² While an increased risk of hip fracture with haemorrhagic stroke has been noted compared with ischaemic stroke,²⁴ there is no evidence that patients with haemorrhagic stroke are at higher risk of perioperative complications particularly if the underlying cause of the haemorrhage has been resolved.²⁵ However, it is reasonable to avoid perioperative hypertension in such patients to prevent a recurrence. In the context of a large stroke which is likely to have poor rehabilitation potential even in the absence of a hip fracture, surgical repair of a hip fracture will have limited functional benefit even if it helps with pain control. In patients who have a stroke-related hemiplegia ipsilateral to the hip fracture, there is concern that surgical repair of the fracture will not result in improved function if there is no corresponding recovery from the stroke. However, such patients will still have the benefit of compensating for their function using the unaffected side. In patients who have a stroke-related hemiplegia contralateral to the hip fracture, the patient potentially loses use of both lower limbs, and any rehabilitation even with surgical correction of the hip fracture is likely to prove even more difficult as there is limited likelihood of the fractured limb compensating for the hemiplegic limb and vice versa. In our patient discussed here, he had reasonably good premorbid function and was ambulant independently at baseline, making him an optimal candidate for operative intervention despite the increased risks associated with stroke. Furthermore, our patient suffered a small stroke with limited impact on physical function. Palliative and non-operative management may also be reasonable in patients who have other life-limiting comorbidities, particularly in patients who show signs often correlated with imminent death, such as Cheyne-Stokes breathing, decreased level of consciousness and decreased urinary output.²² Scores such as the Nottingham Hip Fracture Score and Almelo Hip Fracture Score are useful in determining the risk of early mortality following hip fracture surgery in frail elderly patients.^{26 27} Another consideration that has not been highlighted is to explore the adequacy of social support for our patients. A patient who has a strong support network may opt for non-operative management with the understanding that their family will be able to provide for them without significant implications on their quality of life. It has been postulated that the higher incidence of non-operative management of hip fracture in Singapore is contributed by the presence of good family support for most elderly patients.²¹ In the case discussed here, there were noticeable limitations to his long-term care plan. Our patient opted for operative management despite the high risk because he understood that it will be a very difficult and lengthy process for him to return to self-care without surgery, and he had no other caregivers available to look after him should he not regain functional independence.

The Scottish Intercollegiate Guidelines Network (SIGN) and the UK National Institute for Health and Care Excellence advocate for early surgical repair of hip fractures within 48 hours of admission, as early surgical intervention has been found to be associated with reduced pain, decreased length of hospitalisation, return to independent living and reduced mortality.^28–33 However, previous perioperative stroke guidelines suggest that non-urgent, non-essential surgery be deferred in the context of an acute stroke. Early reviews suggested deferring non-cardiac surgery for at least 2 weeks to allow for recovery of cerebral autoregulation, and in the event of a large ischaemic stroke (greater than one-third the distribution of the middle cerebral artery), it is advisable to defer surgery for at least 1 month.³⁴ Later guidelines suggest that non-urgent surgery should be delayed for at least 3 months to allow for recovery of cerebral autoregulation, with individualised considerations for earlier intervention in urgent surgeries after a careful consideration of risks and benefits.³⁵ A recent study by Jørgensen et al suggests that a history of stroke was associated with adverse outcomes following surgery, especially if the time between stroke and surgery was less than 9 months.¹⁸ Even after 9 months, there was still an increased risk of adverse outcomes, but the risk appeared stable subsequently.¹⁸ Regardless of the delay, patients with previous stroke will still have significantly poorer postsurgical outcomes, with an increased incidence of sepsis, renal failure, myocardial infarction and double the risk of immediate postoperative mortality.³⁶

The application of such recommendations to patients who present with concomitant stroke and hip fracture remains difficult. Hip fracture repair is not typically considered a non-urgent surgery, so it is questionable whether the guidelines dictating deferment of non-urgent, non-cardiac surgery after an acute stroke apply to hip fractures. While not an absolute emergency, guidelines consider hip fracture repair to be an urgent procedure due to improved outcomes with minimisation of delays. SIGN guidelines suggest that

Patients should be fully evaluated before hip surgery, and short unavoidable delays to gain improvement in reversible medical conditions such as restoration of circulatory volume, correction of anaemia and optimisation of electrolytes can and should be undertaken before surgery. However, chasing unrealistic medical goals should not lead to delay. Surgery should not be delayed for infective pulmonary conditions as improvement is unlikely in the presence of continued immobility and pain.²⁹

The same argument could be made for patients who had a concomitant stroke. The risk of recurrent stroke is not likely to be ameliorated with any insubstantial delay. On the contrary, delay in the correction of a hip fracture is likely to have significant impact on stroke rehabilitation in view of persistent immobility. Expedition of rehabilitation after stroke (within the first 24–48 hours) is important for improving poststroke outcomes.³⁷ Furthermore, a substantial delay may also have implications on the feasibility of hip surgery at a later date. For intertrochanteric fractures, soft callus formation and partial union can be expected at 6 weeks, and hence the surgical reduction in fracture could prove more difficult. For neck of femur fractures, delays in surgery beyond 3 months could still lead to the development of muscular contractures that make the surgery more difficult. Furthermore, the presence of a stroke is likely to exacerbate the likelihood of contractures with continued immobility, with poststroke contractures typically developing within 6–12 weeks after a stroke.^{38 39}

The decision between operative and non-operative management typically lies within the responsibility of the orthopaedic surgeon.²² However, in a patient who presents with concomitant stroke and hip fracture, multiple disciplines are involved in decision-making. Inconsistent input from different disciplines leads to confusion for patients and their families when trying to make a decision regarding the risk and benefit of surgery, as well as the best time frame within which to proceed with surgery. It may be prudent to consolidate the multiple considerations in a multidisciplinary team meeting. Without consistent guidelines on the means to weigh the risks and benefits of surgery in such patients and without a consistent team advocating for the patient, there is concern that there will be a default to non-operative management and subsequent institutionalisation for patients who would have otherwise benefited from surgery.

Learning points.

• Concomitant presentation of stroke and hip fracture is not common, but there is a lack of consensus on the ideal treatment trajectory for these patients.

• There is argument to be made for early surgical correction of hip fracture in suitable patients with concomitant stroke, taking into consideration factors such as functional status, comorbidities, type and severity of stroke, and social support.

• Further research needs to be done in terms of risk and benefit stratification for such patients to determine whether and when surgical intervention should be carried out.

• A multidisciplinary orthogeriatric team will need to consider these factors and advocate for surgery in suitable patients.