Abstract
Hip fractures in the elderly individuals are a complex problem. Our objective was to determine whether orthogeriatric treatment is effective in terms of reducing length of hospital stay, morbidity, and mortality of elderly patients with a hip fracture compared with orthopedic (traditional) treatment. From July 2009 to May 2011, patients older than 65 years with a hip fracture were followed prospectively. They were co-treated by geriatric and orthopedic teams. This cohort was compared with a retrospective cohort followed from January 2007 to June 2009 that was managed by the orthopedic surgery team only. Epidemiology, pre- and postoperative hematocrit, and renal function were registered. Also, in-hospital and distant mortality data (determined by consulting the national registry), mortality-associated factors, postoperative complications, hospital stay length, and transfers to other services were registered. One hundred and eighty-three patients in the retrospective group and 92 in the prospective group were included in this study with a median follow-up of 26 months (interquartile range: 13-41). The average age was 84 years and 74% of patients were female. Intertrochanteric fracture accounted for 51% of the cases. There was no difference between groups with regard to hospital stay length, hematocrit at discharge, in-hospital mortality, long-term survival, or transfers to internal medicine or the intensive care unit. It did show differences in the transfer to the intermediate care unit, prolonged hospitalizations (>20 days), and diagnosis of delirium and anemia requiring transfusion. In the present study, orthogeriatric treatment is slightly more effective than traditional treatment in terms of morbidity, but there is no difference in hospital stay length or mortality. Further studies and longer follow-up are needed to draw more conclusions.
Introduction
Age-related complication, such as hip fracture, is of considerable epidemiological importance. 1 In the United States, the annual number of hip fractures is estimated as 250 000 and it is projected that this figure will double by 2040. 2 The costs associated with hip fracture, including acute care, rehabilitation, and home care exceed $5 billion annually. 3
Chile is now a society in demographic transition because of increasing life expectancy and declining fertility rate. In the 2002 census, the population older than 60 years reached 11.4% (1 717 478 people). A review of hospital discharges across the country showed an overall incidence of hip fracture of 23.5 per 100 000 inhabitants, 18.8 and 28.2 in men and women, respectively. In women older than 75 years, the rate increases to 617 cases per 100 000 inhabitants, making this group the one with the highest incidence of this disease. The average age of occurrence is 72.5 years for males and 75.6 years for females. 4
Risk of death from a hip fracture is 4% in the immediate postoperative period and 25% at 1 year. In addition, 25% of patients with a hip fracture will be institutionalized at discharge and only a half will be able to recover their previous functional status. 5
Considering the data presented above, there is a need for comprehensive management of patients with hip fracture. During the 60s, there appeared in England the first comanaged unit of orthopedic surgeons and geriatricians (the orthogeriatric team). In 2009, an orthogeriatric team started to treat patients with hip fracture in the author’s hospital.
The aim of this study was to compare the length of hospital stay, morbidity, and mortality between orthogeriatric and traditional management (only orthopedic surgeons) of patients with a hip fracture.
Materials and Methods
Upon approval by the ethics committee of the author’s hospital, a comparison was conducted since July 2009 between a retrospective cohort (control group) and a prospective cohort (intervention group) of hip fractures in patients older than 65 years. The retrospective cohort was obtained from the database of the author’s hospital and included all patients admitted to the hospital between January 2007 and June 2009 with a hip fracture diagnosis. Inclusion criteria were patients older than 65 years, hip fracture (any type), and acceptance of informed consent (patient or representative). Exclusion criteria were patient transferred from another health center, periprosthetic fractures, and pathological fractures.
Medical records and x-rays were reviewed. This group was handled in the traditional way, only by orthopedic surgeons.
The intervention group was followed prospectively from July 2009 to May 2011. This group met the same inclusion criteria mentioned for the retrospective group. In this study group, all patients received an assessment by the geriatric team at admission, daily during hospital stay, and as outpatients at 3, 6, and 12 months to assess their general condition, basic activities of daily living, and adherence of calcium–vitamin D and bisphosphonates prescription.
Exclusion criteria were patient transferred from another hospital, periprosthetic fractures, pathological fractures, and lack of assessment after hospital admission by the geriatric team.
The primary outcome was length of hospital stay. Secondary outcomes were as follows: in-hospital mortality and long-term survival, transfers to other services (internal medicine, intermediate care unit, and intensive care unit [ICU]), admission and discharge hematocrit and renal function examination, and postoperative complications.
The following data were recorded: Epidemiological data Characteristics of the fracture and the affected side Laboratory tests: hematocrit and renal function (ratio of blood urea nitrogen/creatinine) at admission and discharge In-hospital complications: deep venous thrombosis (DVT), pulmonary embolism (PE), urinary tract infection (UTI), anemia requiring transfusion (hematocrit <30), delirium, and wound infection Hospital stay length: days from admission to surgery and surgery to discharge and total stay length Transfer to other services (internal medicine, intermediate care unit, and ICU) Prescriptions at discharge for thromboprophylaxis and calcium + vitamin D Mortality and its causes (checked in the national health registry)
Statistical Analysis
Assuming an expected mean hospital stay of 9 days in the control group, a mean hospital stay of 11 days in the intervention group, and a standard deviation of 4 days for both groups, an estimated sample size of 85 patients per group was calculated in order to achieve 90% power at standard significance levels (.05).
First, descriptive statistic tests were performed to evaluate the characteristics of the studied patients. The Fisher exact test was used to evaluate univariate association of categorical variables. Quantitative variables were compared using Mann-Whitney or Student t tests according to data distribution characteristics and variances, which were tested by the Kolmogorov-Smirnov and Levene tests, respectively. Kaplan-Meier curves compared with the log-rank tests were used to evaluate long-term survival. The analysis was performed by a statistician who was unaware of the clinical evaluation process using Stata v10.0 (StataCorp, 1996-2011)
Results
There were a total of 275 patients recruited, 183 in the retrospective group and 92 in the prospective group. Median follow-ups were 39 month (interquartile range [IQR]: 27-45) and 11 months (IQR: 6-16) for the control and intervention groups, respectively, with a median of 26 months (IQR: 13-41) for both groups. The most common fracture was the pertrochanteric type in both groups, with an average age of 84 (Table 1).
Patient and Laboratory Characteristics
a Fisher exact test.
Regarding laboratory tests (hematocrit and renal function), no significant differences were found between groups.
Regarding our primary outcome, there was no difference in total hospital stay length (Figure 1), days from patient admission to surgery and from surgery to discharge (Table 2). There were significantly fewer prolonged hospitalizations (>20 days; P = .039; Table 2) and more transfers to intermediate care units in the prospective group (P < .001; Table 2).
Hospital stay according to study group.
Complications, Services Transfers, and Hospital Stay
Abbreviation: IQR, interquartile range.
a Fisher exact test.
b Log-rank test.
c Mann-Whitney U test.
The most frequent postoperative complications were delirium. In the prospective group, there was a higher detection of delirium and anemia requiring transfusion (both P < .001). When analyzing the influence of delirium in the prospective group, it was observed that the median length of stay in patients without delirium was 7 days (IQR: 6-9) versus 9 days (IQR: 7-13) in patients with delirium. Therefore, delirium was associated with statistically significantly longer hospital stay in the prospective group (Mann-Whitney U, P < .001). There were no differences in terms of surgical site infection, PE, UTI, or DVT. Hospital mortality for the control and intervention groups was 1.1% and 2.2%, respectively (P = .46). Long-term survival was not different between groups (log-rank test 0.9) having an 87% of survival rate at 1 year follow-up (Figure 2). Mortality causes did not differ between groups. Cardiopulmonary disease was the primary cause of mortality and cancer was the second most common cause.
Patient survival according to study groups.
Finally, regarding outpatient prescriptions, all patients received calcium and vitamin D in the intervention group compared to only 5% in the control group. In both groups, the prescription for thromboprophylaxis was 100%.
Discussion
This study reflects the first Chilean comparison of orthogeriatric versus traditional treatment in patients with hip fracture. The orthogeriatric program consists of 5 geriatricians and 2 residents who are called when a patient with fracture is admitted to the hospital. At admission, patient’s health condition, basic and advanced activities of daily living performance, and family assessment are registered. The geriatricians advice the orthopedic team on the best surgical intervention timing. Postoperatively, they are in charge of the patient’s general health care, while the orthopedic surgeon is in charge mainly of the rehabilitation. As outpatients, at 3, 6, 12 weeks, and every 3 months until 1 year postop, they assess their general condition, basic activities of daily living performance, and adherence to thromboprophylaxis, calcium–vitamin D and bisphosphonates prescription.
The results of this study show a small benefit of geriatric intervention for hip fracture in the elderly patients. The benefit is demonstrated mainly in improved detection of medical–geriatric disorders. Specifically, it increases the diagnosis (and probably improves the prevention and treatment) of medical conditions such as anemia requiring transfusion and delirium. Even though transfer to intermediate care units was increased, this did not translate into longer hospital stays, but probably in better management of medical conditions.
This study also shows the lack of diagnosis, management, and/or registration of medical conditions in orthopedic surgery. This is evident when assessing calcium + vitamin D prescription at hospital discharge and delirium diagnosis. There is a 19% diagnosis of delirium in the control group as compared to 60% in the intervention group. It is important to note that hospital stay was longer in patients who had delirium. That is why it is believed that the presence of a geriatrician could eventually shorten hospital stays. This finding is consistent with that reported in the literature 6,7 but not found in this study (primary outcome).
Differences in transfusion quantity between groups are not clearly explained because there is no difference in admission or discharge hematocrit. One possible explanation is that patients with mild anemia received a transfusion mainly in the orthogeriatric group and not in the orthopedic group. In cases of mild anemia geriatricians, only transfuse 1 unit of red blood cells, quantity that could be insufficient to significantly improve hematocrit.
International literature also shows that delirium is an independent predictor of increased mortality in hospitalized elderly patients. 8 Therefore, although not identified in this study, it is probable that there will be a difference in mortality with longer follow-up.
Regarding study weaknesses and areas of possible bias, one of our study groups is a retrospective cohort. The hospital ethics committee recommended the authors not to have a prospective control group, given that there was enough evidence of the benefits of an orthogeriatric intervention. Therefore, a retrospective control group was taken.
As mentioned, this study is also limited by the lack of power, leading to type II errors (eg, mortality); however, it has enough power to show a difference in hospital stay length (primary outcome). Other limitation is that it has a short follow-up (follow-up of prospective group is 11 month).
One of the strengths of this study is that it adequately reflects the type of hip fracture patient and the better diagnosis (and probably better treatment) of medical conditions in the orthogeriatric group and the absence of a longer hospital stay. The latter is a very important issue because it is a surgeon’s general belief that internal medicine specialists prolong, sometimes unnecessarily, hospital stay.
Also this is the first Chilean comparison of orthogeriatric versus traditional treatment in patients with hip fracture.
Taking into consideration the elderly polypathology, polypharmacy, high risk of developing medical complications (renal failure, anemia, thromboembolism, constipation, delirium, etc), the authors currently recommend the management of these patients in units where there is a co-treatment between orthopedic surgeons and geriatricians. 9 –11 Since 2009, the author’s hospital has developed an orthogeriatric program. Future studies with functional outcome and longer follow-up are required to draw more conclusions.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.