Fractures of the hip, particularly those occurring in the elderly, are medical, social, and economic challenges. The number of hip fractures occurring each year in an older and frailer population is increasing. Currently, approximately 250,000 hip fractures occur annually in the United States, at a cost of approximately $9 billion. With the projected growth of the elderly population, the incidence is expected to double by 2050.5,6
Femoral neck and intertrochanteric fractures occur at approximately equal frequencies, and may be secondary to low-energy or high-energy mechanisms. These fractures most commonly occur in elderly individuals as a result of low-energy trauma, such as after a fall from standing. Women who sustain intertrochanteric fractures are more likely to be older, more dependent in activities of daily living, and typically are home ambulators before their hip fracture compared with women who sustained femoral neck fractures.5,12
Most of these fractures will require operative intervention to restore patient mobility and optimize the postinjury level of function. Advances in medical treatment, anesthetic techniques, and implant technology have resulted in improved fracture fixation and restoration of postoperative function with decreased rates of complications. Continued improvements are needed, however, to additionally improve patient outcomes. Because resources are becoming increasingly limited and their use is becoming scrutinized more closely, functional outcome measures have been used increasingly to examine cost-effective intervention strategies for the clinical problems encountered.
At our institution, before 1990 the clinical course for patients admitted with hip fractures was variable and depended largely on the individual preferences of the admitting surgeons. However, as our knowledge of perioperative care of the patient with a hip fracture increased, the need for standardization of care to provide efficacious, cost-effective treatment led to the development of a clinical pathway. The purpose of this paper is to describe the clinical pathway used since 1990 at the Hospital for Joint diseases and report the effect of this pathway on patient outcomes in a series of elderly patients with hip fractures. (Appendix 1).
Admission and Preoperative Evaluation
Evaluation of the patient who sustains a hip fracture typically is initiated in the emergency department. Because the majority of hip fractures occur in the elderly population, a detailed history is necessary to determine the mechanism of injury, comorbid conditions, and medications. Elderly patients who live alone frequently experience delays of hours to days before presentation to the emergency department; such patients may be dehydrated, delirious, and nutrionally deprived. Therefore, the initial assessment should include an assessment of hemodynamic stability, vital signs, and urine output.
The clinical deformity of the involved lower extremity will vary depending on the fracture “personality.” Nondisplaced fractures may present with virtually no clinical deformity, whereas displaced, comminuted fractures typically present with the classic shortened and externally rotated extremity. Range of motion of the hip generally is painful and should be avoided. Neurovascular injuries are rare, but a careful evaluation should be done. Preexisting peripheral vascular disease, or peripheral neuropathy requires careful monitoring of the skin and avoidance of excessive pressure during reduction maneuvers. Evidence of preexisting sacral or heel decubiti should be noted and recorded. Hip fractures in the elderly population may be accompanied by other fractures that are prevalent in osteopenic bone, such as vertebral compression and wrist fractures. Therefore, our preoperative assessment includes a detailed physical examination of the entire patient, with special attention given to areas of ecchymosis, swelling, and reported pain.
Radiographs of the hip fracture include AP projections of the pelvis and hip and a cross-table lateral of the proximal femur. If the fracture pattern is unclear, an AP projection of the hip is obtained with axial traction and internal rotation applied to the hip. If plain radiographic assessment reveals no apparent abnormality, but suspicion for hip fracture is great (groin pain on hip motion or axial load), an MRI scan is obtained to determine whether an occult hip fracture is present. The MRI scan is obtained while the patient is being evaluated in the emergency department and is used to determine whether the patient will be admitted to the hospital or sent home. Ecchymotic, painful, or swollen areas are evaluated radiographically to rule out associated injuries. A radiograph of the chest is taken in patients in whom surgical treatment is anticipated.
Diagnostic evaluation also includes an electrocardiogram, urinalysis, and basic blood work, including a complete blood cell count, electrolyte levels, and coagulation profile (prothrombin and partial thromboplastin time). Typing and screening are obtained in anticipation of surgery. In patients with cardiopulmonary disease, functional debilitation, or dementia, we do a baseline arterial blood gas (ABG) analysis.
In elderly patients, cardiopulmonary disease is a major determinant affecting the patient’s ability to tolerate prolonged recumbency, to have surgery, and to participate in a rehabilitation program. On admission, we request consultation by internal medicine and anesthesia for preoperative medical optimization and clearance. Appropriate multidisciplinary consultation, as dictated by the patient’s comorbid conditions, is pursued before operative intervention. Patients with preexisting cardiac conditions occasionally require echocardiography to evaluate cardiac function, whereas pulmonary function tests are indicated in patients with underlying lung disorders. Social service evaluation is initiated with the family to assess functional needs, social network and support, financial resources, and options for discharge planning.
On hospital admission, the patient is given an overhead trapeze frame to allow upper extremity assistance in mobility while in bed. A pillow is placed under the affected extremity for comfort. We do not place patients with hip fractures in traction (skeletal or skin), before operative intervention because traction places the hip in relative extension with a consequent decrease in capsular volume. The increase in intracapsular pressure often results in increased hip pain. Injectable narcotic analgesics are administered and are titrated to the patient’s level of pain. Intravenous fluid replacement is initiated with placement of a urinary catheter to monitor urine output. The patient is given an incentive spirometer and instructed on its use, with coughing and deep breathing exercises encouraged. Intermittent, pneumatic impulse boots are placed on both lower extremities, and chemoprophylaxis against deep venous thrombosis is initiated. We prefer the use of injectable low-molecular weight heparin preparations because they have been shown to be efficacious for the prevention of deep venous thrombosis with a relatively low incidence of complication without the need for coagulation profile monitoring.7 Moreover, because injectable low-molecular weight heparin preparations have relatively short half-lives, they can be discontinued 12 hours before surgery without an increase in intraoperative bleeding complications. Although warfarin sodium provides effective prophylaxis against thromboembolic complications, its relative long half-life and difficulties with reversal limit its clinical use for preoperative chemoprophylaxis.7
If operative treatment is indicated, we proceed with surgery as soon as the patient’s condition has been medically optimized and he or she is cleared for surgery. Although the effect of operative delay on mortality after hip fracture is controversial, it seems that an operative delay is appropriate only to permit stabilization of existing medical problems. One study showed that a surgical delay of 3 days approximately doubled the mortality at the first postoperative year when the factors of age, gender, and number of medical comorbidities were controlled.17
If nonoperative treatment is chosen as the definitive mode of treatment, such as in nonambulatory patients with unacceptably high medical risks, appropriate consultation is pursued to medically optimize each patient’s condition. The patient is mobilized out of bed as soon as possible, with appropriate analgesia for patient comfort and prophylaxis against thromboembolic phenomena. If the patient is able to participate in physical therapy, ambulation training with assistive devices and instruction on functional activities are pursued. Nonambulatory patients are log-rolled every 2 hours; such patients warrant close scrutiny for the development of decubitus ulcers, especially over bony prominences such as the sacrum or heel.2 Patient disposition is discussed with the patient and proxies on admission, and formal discharge planning is initiated.
Day of Surgery
In patients in whom operative treatment of the fractured hip has been indicated, we routinely discontinue the use of chemoprophylaxis for deep venous thrombosis 12 hours before the anticipated time of surgery, but continue intermittent, pneumatic impulse boots until the patient is transferred to the operating table. Unless laboratory abnormalities were present on the previous day, we typically do not obtain blood work the morning of surgery.
The patient is kept NPO on intravenous fluid replacement. Essential medications are allowed on the morning of surgery with small sips of water. A careful neurovascular examination is done on the morning of surgery, with documentation of soft tissue status.
We prefer the use of spinal anesthesia for surgical fixation of the hip. After intravenous sedation is given, the patient is rolled onto the affected side and spinal anesthesia is administered in the lateral decubitus position. General anesthesia may be indicated in patients in whom multiple procedures are anticipated during the same anesthetic period (such as concomitant surgical fixation of a distal radius fracture), patients with severe aortic stenosis or coagulopathy in whom spinal anesthesia would be contraindicated, or in patients with dementia or those who are uncooperative who would be unable to comply with spinal anesthesia. The patient then is positioned according to the planned procedure. Internal fixation of displaced femoral neck or intertrochanteric fractures usually is done with the patient supine on a fracture table. Positioning for prosthetic replacement is dictated by the surgical approach at the discretion of the operating surgeon. Prophylactic antibiotics are given before the beginning of the operative procedure. We prefer the use of a first-generation cephalosporin unless the patient had a previous allergy.
At the conclusion of surgery, we assess limb rotation, limb length relative to the contralateral side, distal capillary refill, pulses, and neurosensory status. We routinely obtain final, postoperative radiographs in the operating room. In the recovery room, we obtain postoperative laboratory studies, including a complete blood cell count and basic metabolic profile. An electrocardiogram is obtained and compared with the preoperative electrocardiogram. Patients with histories of cardiopulmonary dysfunction and those with new changes indicated by electrocardiography are placed on rule-out myocardial infarction protocol, with serial serum cardiac isoenzyme analysis and electrocardiograms.
In the recovery room, intermittent, pneumatic impulse boots are placed on both lower extremities. Incentive spirometry and deep breathing exercises are instituted. Injectable low-molecular weight heparin is restarted 12 hours after surgery. Prophylactic antibiotics are continued for 24–48 hours after surgery. The patient is given injectable narcotic analgesics titrated to his or her level of comfort; alternatively, patient-controlled analgesia is used if the patient can cooperate. Intravenous fluids are maintained as the patient’s oral intake is advanced from ice chips to clear liquids to full liquids. The urinary catheter is continued until the patient is able to use a bedpan. Patients at high-risk, such as those who are unable to be extubated or those with significant cardiopulmonary histories, are transferred to a monitored setting (intensive care or postoperative unit) for observation.
Four to 6 hours after surgery, the patient is reexamined. Neurovascular assessment is done, and the patient’s comfort, hemodynamic stability, wound suction drainage, and status of the surgical dressing are documented.
Postoperative Day 1
On the morning after surgery, suction drains, if present, are discontinued if drainage is less than 50 cc for the preceding 8 hours. The patient’s calves are palpated for tenderness or swelling; any abnormalities are evaluated with duplex ultrasonography.
The urinary catheter is discontinued; if the patient is unable to void after 8 hours and catheterization yields greater than 300 cc, the catheter is retained and antibiotic prophylaxis is continued until the catheter is removed. Most patients are able to tolerate a regular diet by the first postoperative day. At the completion of antibiotic prophylaxis, if the patient is able to tolerate a regular diet, intravenous fluids are discontinued with retention of the saline lock.
The patient is mobilized to a chair with incentive spirometry and deep breathing exercises encouraged. We think that an overhead trapeze frame is essential for the patient to be able to self-mobilize and regain functional independence. The forces exerted across the hip when a patient uses his upper extremities to transfer onto a bedpan approach four times body weight.10 An overhead trapeze helps to reduce this force and eases transfers in and out of bed. Physical therapy is used for ambulation and strength training.
Patients whose fractures were treated with internal fixation have no restriction regarding range of hip motion; patients who had prosthetic replacement, however, are limited to less than 90° hip flexion for 6 weeks. In addition, hip adduction and internal rotation are contraindicated if the prosthetic replacement was done from a posterior approach; while in bed, these patients are instructed to maintain hip abduction with a pillow between their legs.
A complete blood cell count and basic metabolic profile are obtained. We consider transfusion of packed red blood cells if the hemoglobin is less than 9 mg/dL and the patient is symptomatic. In patients with preexisting comorbid conditions or advanced age, we have a lower threshold for replacement of blood products. In patients who are to receive warfarin chemoprophylaxis, we monitor the prothrombin time (PT) and partial thromboplastin time (PTT) daily, with titration of the nightly warfarin dose to achieve a therapeutic level of anticoagulation. This is initiated on the first postoperative day.
In patients at high risk or those with a history of cardiac events, based on our myocardial infarction protocol, three negative sets of cardiac isoenzymes and serial electrocardiograms obtained 8 hours apart rule out the presence of a myocardial infarction. We prefer to transfer our patients to a regular room as soon as the medical condition allows (the patient is extubated and hemodynamically stable).
A consultation is made to the rehabilitation service for evaluation for possible in-house rehabilitation. In our experience, most elderly patients who have surgical fixation of hip fractures will require formal, in-house rehabilitation. In addition to determining suitability for inpatient rehabilitation, the physiatrist outlines an individualized plan for daily therapy, including goals for ambulation, negotiation of steps, and functional activities. This often involves a coordinated effort between the physical and occupational therapists.
Postoperative Day 2 and Beyond
Each day the patient’s calves are palpated for tenderness or swelling. If suction drains still are present, they are discontinued. Unless the patient is incontinent of urine, the urinary catheter is discontinued, if still present. After the urinary catheter and wound drains are discontinued and the course of prophylactic antibiotics is completed, the saline lock is removed.
By postoperative Day 2, hip pain usually is well-controlled. Many patients no longer will require injectable narcotic medications and will have progressed to taking oral narcotic analgesics. Almost all patients are tolerating a regular diet by this time. Laxatives and stool softeners are given if the patient is having difficulty with bowel movements. Beginning on postoperative Day 3, the wound dressing is changed daily until the wound is dry.
We monitor the hemogram and basic metabolic profile for at least the first 3 postoperative days. If a patient is taking oral anticoagulation therapy, we monitor the coagulation profile daily until a therapeutic level is achieved. Unless the patient has laboratory abnormalities, we do not routinely check the hematocrit or serum electrolytes after postoperative Day 3.
By the third postoperative day, it usually has been determined whether the patient will require inpatient rehabilitation. Transfer or discharge plans are established and discussed with the patient, family, and multidisciplinary team. Ambulation training and strengthening exercises are done at least twice daily, and the patient is out of bed and in a chair as much as tolerated.
After transfer to the rehabilitation service, usually by postoperative Day 4 or Day 5, patients are followed up at least twice weekly to assess their progress with therapy. Patients with wound drainage are followed up daily until the wound is dry.
For the duration of the hospital stay, prophylaxis against thromboembolic phenomena is continued, with intermittent, pneumatic impulse boots and injectable low-molecular weight heparin (unless the patient is taking oral warfarin sodium). Incentive spirometry is encouraged while the patient is in bed. If the patient is in the hospital by 1 week after surgery, followup radiographs are obtained.
Multidisciplinary conferences are held daily and are coordinated by the case manager or social services. Patient education materials are given to the patient and family, and formal discussions of discharge planning are initiated. If a skilled nursing, subacute, or rehabilitation facility is necessary after discharge, transfer planning is begun and appropriate referrals are made. Alternatively, if the patient is to be discharged to home, arrangements for visiting nurse service and home therapy are made. A well-defined plan for discharge is outlined, including adaptive equipment necessary for functional maximization at home.
Physical therapy is instituted on the first postoperative day. The patient is instructed on transfers in and out of bed, and ambulation training is begun with assistive devices. If the patient is unable to tolerate this transfer, he or she is assisted into a dangling position. The therapist provides instruction in bed mobility, keeping in mind that the surgically treated limb should be kept in proper alignment.12 Early mobilization out of bed after surgery for a hip fracture is important for the general well-being of the patient; it reduces the risk of deep vein thromboembolism, pulmonary complications, skin breakdown, and decline in mental status.8,12 Mobilization also inspires confidence and encourages the patient while he or she is recovering. In our experience, regardless of the weightbearing status prescribed, the elderly patient with a hip fracture self-modulates weightbearing on the affected extremity based on pain perception; therefore, we allow patients who have had hip fractures to bear weight postoperatively as tolerated.8
Physical therapy ambulation goals are 15 feet with moderate assistance on postoperative Day 1, 20 feet with minimal assistance on postoperative Day 2, and 40 feet on postoperative Day 3. On postoperative Day 3 or Day 4, the patient is instructed on stair climbing with supervision. Ambulation training is progressed as tolerated by the patient, with gradual weaning of the walker to forearm crutches, and supervised stair climbing. These patient goals are subject to modification and are individualized to the patient’s physical, psychologic, and social situations.
In addition to ambulation and mobilization, endurance and strength training are initiated. Exercises done with the patient supine include quadriceps sets, heel slides, active assisted hip flexion, active assisted straight leg raising, active hip extension and abduction, and ankle pumps. Quadriceps strengthening is important to facilitate independent transfer ability, whereas hip abductor strength previously has been shown to be significantly correlated with independent ambulation.3 With the patient in the sitting position, knee extension and hip flexion exercises are initiated. Exercises done with the patient standing include straight leg raises, hip abduction, hip flexion, and quarter-knee bends. These are done concentrically with a 5-second isometric hold and then continued eccentrically as the lower extremity is lowered.
As strength increases, exercises are progressed from active-assisted, to active, and then to resistive. Repetitions are increased to enhance the patient’s endurance. Patients whose balance is impaired may require contact guarding when doing exercises while standing. Balance can be improved by ball playing with the patient and by having patients reach for cones while sitting and standing. Deep-breathing exercises and use of an incentive spirometer are encouraged to help prevent pulmonary complications in this vulnerable population.
Functional recovery is maximized by instruction in activities of daily living. The occupational therapist assists the patient in regaining perceptual, motor, and adaptive skills that enhance the patient’s ability to participate in functional activities, such as bathing, toileting, dressing, and cooking. Although a visit to the patient’s home would be ideal for the therapist to evaluate the patient’s functional and adaptive requirements, it frequently is not feasible; it therefore is incumbent on the rehabilitation and social work teams to obtain necessary information regarding the home environment from the patient, family, and friends. Useful information that can significantly enhance patient rehabilitation and adaptive training includes the placement of staircases; access to rooms, especially the kitchen and bathroom; the presence of hazards, such as slippery floors or rugs; and support available to the patient at home. The rehabilitation team then can devise safety precautions that may be taken to minimize patient risk and determine the need for special equipment to optimize functional recovery.
We routinely use adaptive equipment and assistive devices to aid in transfers, ambulation, and activities of daily living. In our experience, although standard walkers provide an excellent support base during stance, they tend to be cumbersome and may be difficult for the elderly to advance. Rolling walkers may allow for easier forward movement, but only should be prescribed for individuals with sufficient coordination and balance to cease forward motion when necessary. Axillary or forearm crutches tend to be less cumbersome than walkers and may be used to provide unilateral or bilateral support. Canes are excellent for maintenance of balance but provide the least joint unloading. However, because they are manipulated easily and are less cumbersome, they generally are well tolerated by the elderly population.
We reserve the use of wheelchairs for patients who are nonambulatory or unable to tolerate weightbearing because of underlying cognitive, neurologic, or medical issues, or in patients with multiple injuries whose functional mobility otherwise would be impaired significantly. We prefer the use of ambulatory assistive devices in patients who are able to tolerate their use to maintain strength, endurance, coordination, and functional independence, although wheelchairs may be used for long distances.
Additional devices helpful in the performance of activities of daily living include long-handled reachers, shoe horns, stocking-aids, and button-fasteners that allow individuals with reduced joint ROM to dress themselves. Shower benches, shower bars, and long-handled sponges help the elderly to bathe in a safe, private manner. Elevated commode seats allow for self-toileting and hygiene. Specialized kitchen utensils with ergonomic grips and other adaptive modifications can aid in independent meal preparation and consumption.
We determine suitability for discharge from the hospital (acute care or rehabilitation) based on several criteria. The patient should be afebrile and hemodynamically stable, and tolerating a regular diet. The patient’s condition should be medically optimized with near-baseline laboratory values. The wound should be clean and dry with evidence of healing. The level of pain should be minimal or well-controlled with oral analgesics. The patient should be able to do basic activities of daily living, use adaptive equipment, and participate in home exercises. Finally, the patient should be cleared by physical and occupational therapists, indicating a basic level of mobility and functional independence.
The comprehensive, multidisciplinary discharge plan is reviewed with the patient and family, with written dates and times of followup appointments and visits from nurses and therapists. The patient or family is asked to verbalize a clear understanding of discharge instructions, limitations, and needs, and agree on the proposed plan. Medications are reviewed with patients and proxies, and prescriptions given.
Between July 1, 1987 and December 31, 2001, 1065 patients were observed by the Geriatric Hip Fracture Research Group at the Hospital for Joint Diseases. To be included in this program, patients had to be 65 years or older, ambulatory before fracture, cognitively intact, living in their own home or apartment, and have sustained a femoral neck or intertrochanteric fracture of nonpathologic origin. All patients were identified at the time of hospital admission and followed up prospectively. Information on prefracture function, ambulatory ability, living situation, and cause and place of fracture was obtained at the time of hospital admission from the patient or through a family member interview. These baseline characteristics have been described in previous papers from our research group.1,11,14 Data on the patients’ hospital procedures were collected by chart abstracting during hospitalization and at patient discharge.
All patients were treated operatively and followed a similar postoperative protocol consisting of early mobilization on postoperative Day 1 with weightbearing ambulation as tolerated. Patients were followed up by one of two trained interviewers at 3, 6, and 12 months after fracture and every 6 months thereafter or until death. If the patient was not available, a family member or caregiver was interviewed.
Before 1990, our hospital did not have a clinical pathway for patients with hip fractures. On January 1, 1990, our hospital initiated the clinical pathway as described in the previous sections. Statistical analysis was done to evaluate the effect of the clinical pathway on patient outcomes by comparing hospital length of stay, discharge status (home or skilled nursing facility), inhospital mortality, and mortality and ambulatory abilities at 6 months and 1 year before and after the implementation of the clinical pathway. Preliminary analyses were done using contingency tables (chi square) or student’s t test or both as deemed appropriate. Probability values < 0.05 were considered significant. To control for confounders, multiple regression analyses were done.
The characteristics of the study population are summarized in Table 1. Three hundred eighteen patients were treated before January 1, 1990 and 747 patients were treated after initiation of the clinical pathway. No significant differences in patient demographics existed between the two groups of patients.
After initiation of this pathway, there were significant decreases in the acute care hospital length of stay (21.6 versus 13.7 days p < 0.001), inhospital mortality (5.3% versus 1.5%), and 1-year mortality (14.1% versus 8.8%) (Table 2). There were no differences in the rate of revision hip surgery, discharge status, or recovery of ambulatory ability between the two groups of patients.
Hip fractures are common injuries in the elderly and are associated with considerable morbidity and mortality. The projected expansion of the geriatric population in the near future will have an additional impact on the healthcare system and caregivers.5,6 Although technical advances in the treatment of hip fractures have resulted in improved fracture fixation and surgical outcomes, a review of our data has suggested that the establishment of a clinical pathway for hip fractures may result in a cost-effective standardization of care, with reductions of acute care length of stay, inhospital length of stay, and 1-year mortality.
Other authors have reported similar findings with respect to the efficacy of clinical pathways after hip fracture and other selected medical conditions. Choong et al reported a series of 111 patients with hip fractures managed either by a clinical pathway (55 patients) or established standard of care (control group, 56 patients).4 Patients managed according to the clinical pathway had a shorter inhospital stay (6.6 versus 8.0 days; p = 0.03). There were no differences in complication or readmission rates between the 2 groups of patients. March et al compared the results after hip fracture between patients treated before (n = 455) and after (n = 481) clinical pathway implementation.9 Implementation of the clinical pathways was associated with increased use of evidence-based best practice and some reduction in acute hospital length of stay, but no notable effect on 4-month mortality or place of residence. Podila et al evaluated use of a clinical care pathway for the management of patients with acute upper or lower nonvariceal gastrointestinal hemorrhage.13 Multivariable regression controlling admission vital signs, comorbid conditions, age, and the etiology of hemorrhage confirmed that admission after pathway implementation was an independent predictor of a reduced length of hospital stay. There were no significant differences in the 30-day rates of recurrent hemorrhage, mortality, or hospital readmission.
The main strength of the current study is that the data were collected prospectively from a large sample of patients, which avoids some of the inaccuracies of chart review found in retrospective studies. Phone followup is somewhat less reliable than direct patient observation; however, studies in this patient group have shown that phone followup is accurate and allows for more efficient followup.15,16
We think that establishment of a clinical pathway for hip fractures may result in a cost-effective standardization of care, maximal use of available resources, and a decrease in the incidence of morbidity and mortality. Importantly, our clinical pathway uses a coordinated, multidisciplinary approach with emphasis on functional recovery and restoration of quality of life.