Predictive Value of Charlson Comorbidity Index and Cofactors on 2-Year Mortality in Older Patients With Intertrochanteric Fractures: A Retrospective Analysis

The aim of this study was to evaluate the role of Charlson comorbidities index (CCI) and cofactors on 2-year mortality in older patients with intertrochanteric fractures. 60 cases with unilateral intertrochanteric fracture were retrospectively analyzed and divided into Low-CCI group (CCI: 1-4) or high-CCI groups (CCI: 5-6). All the patients’ electronic hospital records were reviewed. The preoperative situations (demographic data, comorbidities and fracture conditions), perioperative situations (wait time, operation time, implant choice, blood loss, transfusion or not) and postoperative situations (complications, rst time out of bed, function about 1-/2- week and 2-year mortality) were recorded. 51.67% were in low-CCI group and 48.33% in high-CCI group. The survival rates in low- and high-CCI group were 93.5% and 86.2 % respectively. According to the functional results of 1- or 2- week after operation, no signicant difference was found (P=0.955, 0.140). Log-rank analysis showed that the main prognostic factors were blood loss, rst time out of bed and complication (P<0.05). Multivariate analysis conrmed that complication and rst time out of bed were signicant factor on survival rate (P=0.029, 0.010). Charlson comorbidities index maybe not the indicator of 2-year mortality in older patients with intertrochanteric fractures. In order to improve the prognosis, more attentions should be paid to reduce the complications and encourage postoperative earlier excise out of bed.


Introduction
As a common fracture in elderly people, the intertrochanteric fracture comprises about 8-10% of all fractures 1 . Although fracture is not the direct cause of death, it can induce a progressive deterioration of physical condition and increase the mortality. The mortality of intertrochanteric fractures displays a rising trend due to increased life expectancy and osteoporosis 2 .
Surgery usually supplies an earlier mobilization to reduce complications and mortality, and improves the independent living ability in older patients 3 .To now, surgery at the earliest opportunity has being the main choice for intertrochanteric fractures, but the relative higher mortality is still a major concern to our clinicians. The trauma reaction, physiological decline and underlying diseases are still the important issues requiring our emphasizing 4 . Considering the di culty and complexity, understanding the risk factors of mortality is very important for supplying safer and better treatment.
Charlson comorbidities index (CCI) was a system assigned to weigh the morbidity, which is used to assess the probability of survival 5 . Previous research had con rmed that CCI was an important tool to predict the prognosis (30-, 90-days mortality) of patients with hip fractures 6,7 .Haentjens P et al used lifetable method and found that women and men (older than 80 years old)having a hip fracture often brought them with excess annual mortality at 1, 2, 5 and 10 years after injury 8 . Considering the following deleterious effects of hip fractures, earlier evaluating the risk factors about the mid-to long-term mortality are very important. To our knowledge, the effect of CCI on the mid-to long-term mortality of patients with intertrochanteric fractures has not been studied.
The aim of this present study was to evaluate the role of Charlson comorbidities index (CCI) on 2-year mortality in older patients with intertrochanteric fractures. In addition, we investigated whether other main risk factor in uencing the prognosis.

Materials And Methods
This retrospective study was designed and approved by the Ethics Review Committee of Juntendo University. The informed consents were obtained from the patients in the manuscript. All methods were performed in accordance with the relevant guidelines and regulations. The operations were performed by the same team major in hip. According to the regime of our rehabilitation, the physiotherapist would help patients to mobilize from the second day after operation.
The patients attended rehabilitation twice a day. If possible, they would excise out of bed as early as possible (usually the second day after operation). The following rehabilitation projects mainly based on patients' own physical conditions to accelerate muscle strength, limb coordination and body balance recovery.

Study clinical parameters
Follow-up was de ned from the operation day. 2-year mortality was de ned the survival rate of all enrolled patients at 2 year.
The preoperative situations (demographic data and comorbidities), perioperative situations (wait time, operation time, implant choice, blood loss, transfusion or not) and postoperative situations (complications, rst time out of bed and 2-year mortality)were got from the patients' electronic hospital records. And all the data was manual reviewed independently by two different researchers..

Radiographic evaluation
The fracture classi cation and stability 9 were assessed independently by two different radiologists using PACS system.

Results
All 60 patients got followed-up and bone united uneventfully. Of them, 51.67% were in low-CCI group and 48.33% in high-CCI group. All survival data were available and the survival rates in low-and high-CCI group were 93.5%, 86.2 % respectively ( Figure 1). Three complications were found postoperatively. One patient got urinary tract infection and was cured by antibiotic treatment. Two others had pressure sore, and the wound healed after dressing change. No other secondary displacement, super cial or deep infection was found.

The earlier postoperative functional results of the low-and high-CCI groups
According to the functional results of 1-or 2-week after operation, no signi cant difference was found between low-and high-CCI groups (P=0.955, 0.140, Table 2). The rst time out of bed postoperatively is 1.29±0.79 days.
At 1 week after operation, only 3 patients (5.0%,   Multivariate analysis con rmed that complication and rst time out of bed were signi cant factor on survival rate (P=0.029, 0.010. Table 4).

Discussion
Surgery atthe earliest opportunity has being the main choice for intertrochanteric fractures in older patients. However, the relative higher mortality is still a major concern to our clinicians. CCI was a most widely comorbidity measure method used to assess the probability of survival. Previous research had con rmed that CCI was an important tool to predict the prognosis (30-, 90-days mortality) of patients with hip fractures. What's more, the CCI had been con rmed to predict in-hospital mortality and readmission after orthopaedic surgery accurately 12,13 .
Considering the following deleterious effects of hip fractures, earlier evaluating the risk factors about the mid-to long-term mortality are very important. Lei Jiang et al. retrospective analyzed 1057 hip fracture patients aged above 60 years and found that Charlson comorbidity index was correlated with 5-year mortality after surgery 14 .In this present study, we analyzed the risk factors about 2-year mortality in older patients with intertrochanteric fractures. But no signi cant relationship was found between CCI and mortality rate in this present study. Maybe small sample size and age composition were main in uence factors. The average age of all enrolled patients was 81.7±8.1 years old, and patients older than 80 years old was about 63.9% (39/61).
Gilbert M et al. had got the similar conclusion and thought that high comorbidity index was not related with high morbidity and mortality 15 . But they also pointed out that constrained total hip arthroplasty provided better function to the elderly patients with intertrochanteric fractures. Palanisamy AM et al. compared the functional outcomes in geriatric hip fractures between different implants. The nal results showed that in patients with high CCI, THR usually provided faster recover 16 .Whether CCI combined with other factors can bring better clinical signi cance? It remains to be further studied in future.

Complications
Compared with intracapsular fracture, patients with intertrochanteric fractures usually accompanied with increased blood loss 17 . It's a potential reason for hypotension and hemodynamic disturbance during perioperative period. These uctuations predisposed the higher complication rate. Chen CH et al.
analyzed the risk factors for 1-year mortality in patients with acute hip surgery and strongly recommended that controlling and co-caring postoperative complications could signi cantly reduce the mortality 18 .
In our present study, multivariate analysis con rmed that complication was a signi cant factor on survival rate. Other studies had also got the similar conclusions. FL Zhao et al performed a multivariate analysis and found that BNP, APACHE II score and complications after fracture were reference indexes for predicting perioperative mortality in elderly patients with intertrochanteric fracture 19 .Therefore, it is very important to control complications during perioperative period.

First-time out of bed
Earlier mobilization has been regarded as an important factor for better function after lower limb surgery.
It can reduce complications and improve the independent living ability in older patients 20 .In present study, the mean rst time out of bed postoperatively was 1.29±0.79 days under the guidance of physiotherapist. The earlier weight-bearing time was much shorter than others' reports. Duymus TMet al compared the result of intra-and extramedullary implants in treatment of intertrochanteric fractures and found that proximal femoral nail had better clinical results. The full weight-bearing time was signi cantly reduced to 1.25±0.40 months 21 .
Our multivariate analysis con rmed that rst time out of bed was a signi cant factor on survival rate. And we thought earlier weight-bearing is a better method to reduce the rate of complications and mortalities. Some surgeons worried that earlier weight bearing would bring a higher implant failure rate, but our follow-up study didn't lead to a higher failure rate. In our hospital, the physiotherapists usually help patients weight bearing from the second day after operation if permitted. And the rehabilitation content was adjusted simply according to the actual situation of each patient. Of course, the lower BMI in Japan may have some impact on higher failure rate. To now, no su cient evidence to determine the better weight bearing time. Further studies are needed to be conducted to express it and related risk factors.
At present, there is another view that arthroplasty was a prefer choice for older patients with intertrochanteric fractures. It provided an earlier weight bearing and better functional prognosis 22 .
However, there is still controversial about arthroplasty or internal xation 23 .

Limitations
Some limitations could be found in this study. First, it's a retrospective study with small sample size. The small size reduced the power but provided our own real experience. Second, as a large private hospital, frequently-used hip function scoring standards were not regularly registered. And we used our own semiquantitative based on the patients' electronic hospital records. Third, due to Japanese medical system, not all patients were died in our hospital. The concrete causes of death were still uncertain Fourth, this retrospective study just came from our own hospital, and multicenter and more sample studies are necessary in future.

Conclusion
Therefore, Charlson comorbidities index maybe not the indicator of 2-year mortality in older patients with intertrochanteric fractures. In order to improve the prognosis, more attentions should be paid to reduce the complications and encourage postoperative earlier excise out of bed.

Declarations Data Availability
All data generated or analysed during this study are included in this published article and are available from the corresponding author on reasonable request.