Orthogeriatric co-management for proximal femoral fractures. Can two additions make a big difference?

The socioeconomic burden of proximal femoral fractures is rising with increases in the aging population in modern societies. In Germany, approximately 162,000 proximal femoral fractures were treated in 2017 [1]. The incidence is expected to rise in the future [2] and consequently the costs [3]. Over a lifetime, 16–18% of white women and 5–6% of white men are expected to experience a proximal femoral fracture [4].

In addition to the high incidence and costs, the outcome of these patients is poor. Inpatient complication rates are nearly 20% [5] and the 1-year mortality is 25% [6]. Many patients experience a decline in their performance of activities of daily living (ADL) and they do not regain their pre-fracture mobility level [7]. Consequently, 12 to 15% of patients aged > 65 years move to a nursing home within 6 months of their discharge from the hospital [8].

These findings indicated that there is a need for improvement in the care of proximal femoral fractures in older patients. One approach is to reduce the waiting time for surgery. Current guidelines recommend that surgeons perform surgical treatment for proximal femoral fracture within 24 h of injury, because observational studies have suggested better outcomes can be achieved within this time frame. Earlier surgery was associated with shorter hospital stays, shorter pain durations [9], fewer complications and lower mortality rates [10].

Moreover, due to many comorbidities and an increased age [11] some hospitals combine orthopedic and geriatric expertise in the treatment of these patients. In the traditional care the patients are treated by an orthopedic surgeon and geriatric consultation is only performed on request [12]. Basically, there are 3 different models of orthogeriatric co-management described in the current literature. Firstly, patients can be treated mainly by orthopedic surgeons with routine consultation of a geriatrician. While some studies [13, 14] were able to show an improvement with regard to mortality this effect was not visible in all studies [15]. Secondly, patients can also be treated mainly at a geriatric unit with orthopedic consultation. The meta-analysis of Moyet et al. [16] favors this treatment concept with regard to reducing mortality. In the third mostly performed model the patients are managed together by an orthopedic surgeon and geriatrician. Whether mortality can be reduced with such a model is not clear [16‐19]. However, Rapp et al. demonstrated recently in an evaluation of health insurance data the clear benefit of orthogeriatric co-management with regard to mortality [20]. Additionally, orthogeriatric treatment can result in increased and reduced complication rates [14, 21]. Multidisciplinary treatment is often followed by additional changes in treatment protocols for geriatric patients. These treatment protocols can help to reduce costs, complications and mortality [22‐24]. There is still much uncertainty which of the above mentioned treatment models is most beneficial for this patient collective and how often multidisciplinary treatment is needed [24, 25].

Most previous studies investigated the effects of relatively comprehensive care models with daily interdisciplinary management and changes in geriatric treatment protocols. Often, a major change like that is difficult to implement into clinical practice, whereas the integration of only a few changes can be more feasible. This study aimed to determine whether two additions to the usual orthopedic care could improve process and outcome measurements at the hospital stay and at a follow up of at least 3 months after surgery. The main focus was on a reduction of TTS and complication and mortality rates.

For this retrospective observational study we reviewed the charts of all patients aged ≥70 years that underwent surgery for proximal femoral fractures at our institution (Level I trauma center) between May 2018 and October 2019. Patients with concomitant fractures or injuries that needed treatment and prolonged mobilization were excluded. Further, we excluded all in-patient falls resulting in a proximal femoral fracture. Periprosthetic fractures were excluded. If patients had a second fracture within the study period, one of the admissions was randomly excluded.

This cohort was divided into two groups. The control group received no orthogeriatric co-management (1 May 2018–31 January 2019). The intervention group received multidisciplinary orthogeriatric co-management (1 February 2019–31 October 2019).

Two major changes distinguished the intervention group from the control group:

Both changes were initiated on 1 February 2019. We aimed to improve important outcome parameters with minimal changes (no significant reduction in elective cases, no significant increase in personal resources), compared to the usual care. In the usual care, patients were admitted from the emergency department to our trauma ward. They were visited daily by a resident in trauma surgery. The patients underwent surgery at the earliest time that the operation room and the required staff were available. Consultation was available from the Departments of Internal Medicine and Geriatric Medicine when specially requested. Patients were assessed on the risk of developing a pressure sore and preventive actions were taken, if needed. After surgery, patients received physical therapy on weekdays until discharge and full weight bearing was usually allowed for all patients. Patients with platelet aggregation inhibitors were operated without any change in their medications. Medication with direct oral anticoagulants (DOAC) was stopped for 24 h before surgery and given again at the first postoperative day. Vitamin K anticoagulants were also stopped before surgery and bridged with low molecular heparins. To accelerate the time to surgery, Vitamin K was supplied and the INR was thoroughly checked. A restrictive transfusion regimen was used. All patients with a Hb ≤ 6 g/dl received transfusions. Patients with 6 g/dl < Hb ≤ 8 g/dl received transfusions if symptomatic or by known cardiovascular diseases. Patients with a Hb > 8 g/dl did not receive transfusions. Patients were discharged as soon as they were medical stable and care was ensured. Frequency of physiotherapy, guidelines for mobilization after surgery, prevention of pressure sores, surgical technique, anesthesia, discharge, blood transfusion or bridging of anticoagulation did not change between the groups.

During the time of the study a total of 266 cases aged ≥70 years were identified as having surgery for a proximal femoral fracture at our institution. Two hundred seven met our inclusion criteria and 59 were excluded (Fig. 1). Of these, 103 were treated with the usual care and 104 were treated with the co-management model.

This study showed that two additions in proximal femoral fracture care could significantly reduce the LOS and TTS. Moreover, with these changes, a larger number of patients was satisfied with the treatment. Nevertheless, the changes did not significantly impact mortality or complication rates during the hospital stay or after a 3-month follow-up.

While detecting medical problems and preventing complications is one of the main tasks of the geriatrician in an orthogeriatric setting, Coventry et al. [30] showed higher complication rates after involvement of a geriatrician. This increase is explained by a better detection of complications. A better detection of complications may have prevented this study from showing a reduced complication rate. However, if studies showed reduced complication rates, mostly the least harmful complications were reduced [31, 32]. Delirium is common in geriatric patients after surgery for proximal femoral fracture [32, 33]. When low rates of delirium are presented it has to be questioned whether the hypoactive form of delirium is adequately represented [34, 35].

It has been reported that orthogeriatric co-management can reduce the mortality of geriatric patients with proximal femoral fracture [20]. But there is still controversial discussion which model is most effective [16]. Treating patients on an orthopedic ward with routine consultation of a geriatrician, as performed in this study, has already been found effective with regard to reducing mortality [13, 14]. However, Moyet et al. [16] could only show a slightly and not significant improved mortality for this co-management model. It has to be questioned if treating patients at a geriatric ward with orthopedic consultation would not be more effective. Besides the setting of the co-management, the frequency of consultation seems to play an important role. Only one consultation per week, as in our study, seems to be not enough for reducing mortality [36], whereas daily consultation was able to show improvements [13]. Integrating orthogeriatric care can be accompanied by a learning phase in which the outcome may be worse than before [37]. Such a transition phase may contribute to our findings.

Not only the orthogeriatric co-management model affects mortality and complication rates, but also other process aspects. A reduction of TTS and LOS is a common finding after implementation of orthogeriatric treatment [38]. With the implementation of one reserved operation slot per day we aimed to reduce the TTS and consequently the mortality and complication rates [9, 10]. While we managed to reduce the TTS, we were not able to show a reduced mortality, as expected. The association between TTS and mortality is discussed controversially. Some studies discuss cut-offs that are higher than our TTS in the control group for a better outcome [39, 40]. Reducing the TTS further, as in our study, would therefore be without effect. Other studies indicate that the shown reduction in mortality after earlier surgery is mainly explained by reduced morbidity of these patients [41, 42].

Other treatment aspects of proximal femoral fractures also affect mortality. Knobe et al. [24] showed that the introduction of a special geriatric treatment pathway alone can improve patient outcome for geriatric patients. The best anesthesia technique for proximal femoral fractures is discussed in recent literature and varies extremely between different countries [43]. Yet, there seems to be no clear evidence favoring general or regional anesthesia according to a recent Cochrane review with regard to mortality [44]. Early mobilization and full weight bearing should be performed after surgery for proximal femoral fracture as it increases functional recovery and reduces complication and mortality rates [45‐48]. Latest studies show that liberal transfusion regimens are associated with higher cardiovascular complications in patients with proximal femoral fracture [49, 50]. In our hospital we performed already a restrictive patient blood management as guidelines recommend it. In this study, only the age determined whether a patient was treated interdisciplinary or not. Consequently, patients were included without need for geriatric co-management. Using a test like the ISAR screening tool [51] would enable hospitals to concentrate on patients with need for geriatric co-management. With a more focused patient selection and implementation of geriatric treatment protocols other studies might be able to show improved mortality or complication rates.

Moreover, staff education is another element of orthogeriatric programs [52]. The staff of our orthopedic unit might have improved their knowledge about the diagnosis, prevention and therapy for geriatric syndromes. This gain in knowledge could have improved the therapy for geriatric patients even on days when no geriatrician was involved.

We showed that patients were significantly more satisfied with co-management care. The reasons may be a reduced waiting period for surgery and the more comprehensive care. Boylan et al. showed that higher patient satisfaction is associated with reduced time to surgery and mortality of geriatric femoral neck fracture patients [53]. If we could show increased satisfaction with a less comprehensive model, then satisfaction might also be increased with models that include more frequent geriatric involvement.

Our co-management model reduced the LOS and waiting time for surgery. Additionally, it increased patient satisfaction with the care received. Nevertheless, it could not reduce mortality or complication rates during the hospital stay or at a follow-up of at least 3 months. To achieve the latter effects, a more comprehensive model may be needed.