Hip fracture (HF) is a very common problem throughout the world (1). In elderly individuals it often leads to deterioration in their general condition (2–4). The risk of death in the first year after HF is threefold that of the same age group without HF (5). Many patients experience a decline in physical, social, and mental function and only half of the patients succeed in regaining their pre-HF functional capacity (6, 7). The recuperation period often is accompanied by emotional and social crises, in some cases leading to suicidal ideation (8).

Pelvic fractures (PF) are usually associated with high-energy trauma, are complex, difficult to treat and, at times, have a poor prognosis. They are considered the major cause of mortality among types of skeletal traumas (9). PF are also a common problem in the adult population (10–14).

The rate of PF among all low-energy fractures patients has increased over the years (10, 11, 13). While the rate of HF is on the decline, there is an opposite trend for PF. A nationwide register study from Sweden showed a decline of 18% in the incidence of HF and an increase of 25% in the incidence of PF over the last 16 years (15).

A large proportion of PF patients (11–53%) develop complications associated with PF over the course of hospitalization or after discharge (10, 15–21), with limited mobility (9, 12, 22–25), increased long-term care (LTC) (10, 12, 17, 19, 23, 24), decreased quality of life (16), increased utilization of healthcare system resources (14), and a substantial increase in mortality (9–12, 14–18, 20–24, 26–29).

Compared to the literature on HF, the literature on PF is scant even though the outcomes of the two types of fractures, for example LTC and mortality, are similar (30). Only a few studies, investigated predictors of mortality following PF: age (11, 16, 27, 31), male sex (10, 11, 21), pre-fracture mobility (18, 21), co-morbidity (10, 12, 27), and type or number of PFs are predictors of death (10, 16).

The literature on rehabilitation outcome following HF is copious. In contrast, the literature on rehabilitation following PF is scant. Keene et al. (32) conducted a systematic review to evaluate clinical trials in patients with a fragility fracture of the pelvis or lower limb. They identified 37 studies, none of which was related to PF. Marrinan et al. (25) followed 110 patients who were hospitalized with PF and found that older age and lengthy hospital stay were independent predictors that a patient would need LTC at discharge from the hospital. Pfeiffer et al. (33) investigated the effect on rehabilitation outcomes of interventions aimed at reducing the fear of falling and increasing physical activity. Only 17 of the 115 study participants had PF. Sherrington et al. (34) evaluated the effect of exercise on limited motility and fall prevention in patients with fractures in their lower limbs and PF. They did not assess or report on other variables that could affect the outcome of rehabilitation. Only 31 of 336 study patients had PF and only 15 of the 336 began intervention less than 3 months after the fracture. To our knowledge, no studies have been published that specifically assessed the results of rehabilitation after PF. The present study had two main objectives: (a) to characterize patients who underwent intensive rehabilitation following PF and to compare them to patients who underwent the same treatment following HF and, (b) to identify predictors of successful rehabilitation following PF.

The study population included 144 PF patients and 138 HF patients. As stated above, the HF group has been described in the past (35), and is presented here only as a comparison group for the PF sample. In short, the mean age of the HF patients was 82.5 ± 7.1 years and 107 (77.5%) were women. Forty two (30.4%) had intracapsular fractures and 96 (69.6%) had other types of extracapsular HF. Ninety five (68.8%) underwent internal fixation, 21 (15.2%) total hip replacement, and 22 (15.9%) hemiarthroplasty.

Table 1 summarizes data for the two groups. The mean age of PF patients was 81.5 ± 6.9 compared to 82.5 ± 7.1 for the HF group (p = 0.23). The percentage of women was significantly higher in th PF group (90.3% vs. 77.5%, p < 0.001) and significantly fewer in that group required nursing care prior to the fracture (60.4% vs. 92.0%, p < 0.001). The HF group had significantly higher co-morbidity levels than the PF group based on the Charlson Co-morbidity Index, total score (CCI TS) (2.1 ± 1.9 vs. 1.6 ± 1.7, p = 0.013) and the Charlson Co-morbidity Index, total cumulative score (CCI TCS) (5.9 ± 1.9 vs. 5.3 ± 1.8, p = 0.003). Patients in the HF group were taking significantly more drugs prior to admission to the department than PF patients (10.2 ± 3.3 vs. 6.8 ± 3.7, p < 0.0001). Compared to the PF group, patients in the HF group were at a significantly higher risk to develop pressure sores based on the Norton index (14.1 ± 1.3 vs. 13.6 ± 1.5, p = 0.003).

Compared to the HF group, patients in the PF group had significantly lower creatinine levels (1.1 ± 0.9 vs. 1.4 ± 1.5, p = 0.0003), lower blood urea levels (54.5 ± 33.3 vs. 72.9 ± 37.8, p < 0.0001), and higher albumin levels (3.6 ± 0.5 vs. 3.0 ± 0.4, p < 0.0001).

The complications rate was significantly higher in the HF group. The most prevalent complications were delirium (21.7% vs. 8.3%, p = 0.0016), infection (29.0% vs. 11.1%, p = 0.0002), and cardiovascular (23.9% vs. 5.6%, p < 0.0001).

There were no statistically significant differences between the groups in the other socio-demographic, clinical (MMSE, depression rate or BMI), or laboratory variables. Data on functional capacity, the length of hospitalization in the Geriatrics Department, the rehabilitation outcome, and mortality are presented in Table 2.

There were no differences between the groups in anFIM, FIMa, and FIMd. The mean duration of rehabilitation was significantly longer in the HF group (20.9 ± 7.5 vs. 18.2 ± 7.7 days, p = 0.0007). The outcome of rehabilitation was similar in the two groups with 87 of 144 patients in the PF group (60.4%) and 89 of 138 in the HF group (64.5%) reaching an MRFS-R score ≥ 50 (p = 0.483).

The vast majority of the patients were discharged to their homes at the end of the rehabilitation process in the Geriatrics Department. Only five of 138 patients with HF (3.6%) and five (3.5%) of PF patients were transferred from the Geriatric Department to LTC (p = 0.999).

There was a statistically significant difference between the groups in one-year mortality with 19 deaths (13.8%) in the HF group compared to 9 (6.3%) in the PF group (p = 0.037).

Table 3 shows a comparison between PF patients with a successful rehabilitation and those whose rehabilitation was unsuccessful. There was a borderline difference in age with a mean age of 80.6 ± 7.4 in the successful group versus 82.9 ± 6.0 in the others (p = 0.055). The mean MMSE score was significantly higher in the successful group (24.2 ± 4.5 vs. 19.2 ± 5.3, p < 0.0001) and the rate of infections in the Geriatrics Department was significantly lower (5.7% vs. 19.3%, p = 0.015) in that group. There were no significant differences between the groups in the other variables.

A logistic regression model was built to identify independent variables that were significantly associated with successful rehabilitation (Table 4). The model included sex, CCI TS (because its univariate p value was below 0.1 (p = 0.081) and because it reflects co-morbidity), patient age, and the other variables that reached statistical significance in the univariate analyses. The latter included MMSE and the rate of infection during hospitalization in the Geriatrics Department. In the model (R² = 0.273), only MMSE remained statistically significant at OR = 1.2 (95% CI 1.1–1.3, p < 0.0001).

The results of this study show that there are differences between patients who underwent intensive rehabilitation in the Geriatrics Department for PF and those who underwent it for HF. As expected, most of the patients in both groups were women, but the percentage of women was significantly higher in the PF group. The high rate of women with PF is well documented in the literature with a rate higher than 80% in most studies in elderly populations (15, 18–20, 24, 25, 28). In a Swedish study (43) of over 400,000 patients following osteoporotic fractures, the investigators found that the rate of women among PF patients was higher than among HP patients, but that the differences were lower than in the present study.

In the present study the HF group had a more complex medical history with a higher CCI score, higher rate of medication use, a lower Norton score, and lower kidney function and albumin levels, compared to the PF group. The complication rate was also higher in the HF population. Some differences in the parameters between the PF and HF groups may be attributed to the fact that the HF group was admitted to the Geriatric Department after undergoing surgery and spending several days in the Orthopedic Division, rather than being directly admitted from the Emergency Department, as was the case with the PF patients. This might partially explain the lower kidney function on admission in the HF group, and a higher frequency of complications compared to the PF group. Although the anamnestic pre-fracture functional capacity was similar in the two groups the need for nursing care was higher in the HF patients, a finding that could reflect a higher level of frailty among HF patients. The combination of a high level of frailty and medical complexity in this population could explain the need for a longer rehabilitation period in the HF group to attain a similar functional level. The mortality rate during the first follow-up year was also higher in the HF group. In the study by Lundin et al. (43), the mortality rate over the first post-fracture year in patients over the age of 50 was 25% after HF and 21% after PF. However, another study (12) found that the higher mortality rates among HF patients evened out with PF at 5 years post-fracture. The higher mortality rates over the first year in the HF group could be related to the higher rate of men in this population. Lundin et al. (43) also found that the differences in mortality rates between men and women were higher in the HF group than among PF patients. The burden of illness in the HF group could also explain the difference in mortality. In any event, the mortality rate in the present study over the first post-fracture year among PF patients at 6.3% was lower than in previous studies. In the Swedish study cited above (43), the mortality rate in the PF group was 21% after 1 year. In other studies the mortality rate among patients hospitalized after PF ranged from 9.5 to 27% (15, 18, 23, 28). In a comparison of the PF population in the present study to a previous study of PF patients hospitalized in a Geriatrics Department (20), in addition to the lower mortality rate (6.3% vs. 12.5% in the present study), the infection rate during hospitalization (11.1% vs. 36.7%), and the duration of hospitalization in the department (18.2 vs. 45 days) were also lower in the present population. In a review of the literature we did not find any paper that related specifically to a population of PF patients that were hospitalized for intensive rehabilitation. This may represent a selection bias in the sample of PF patients who were admitted to our department for rehabilitation.

Another aim of the present study was to identify predictors of successful rehabilitation after PF. The present study included 83 patients with Type A PF and 61 with Type B PF. There were no patients with Type C PF. There was no difference between the groups in the outcome of rehabilitation, with 57.8% of patients with Type A PF and 63.9% of patients with Type B PF reaching MRFS≥50. There are several possible explanations for this result. While Type A PFs are stable, stability from the vertical point of view (9) in Type B PF may be sufficient for successful rehabilitation. Of course, the sample size and the retrospective nature of the study could mean that other confounders may be responsible for the lack of difference in rehabilitation between the two groups.

Cognitive state was an independent predictor of rehabilitation success, with every addition of one point in the MMSE score increasing the patient’s chance of reaching MRFS-R ≥ 50 by 20.5%. The association between cognitive state and the outcome of rehabilitation after HF has been reported in many studies in the field (44–49). As noted, no studies assessed the association between cognitive state the outcome of rehabilitation after PF.

In contrast to previous studies of patients who underwent intensive rehabilitation in our department after HF, i.e., with the same staff and identical work routine, factors such as serum albumin level (35) or comorbidity (42) were not found to predict rehabilitation success in PF patients.

The present study has several advantages. It is the first study, to our knowledge, to compare the outcome of intensive rehabilitation among patients with HF and PF, and it is the first attempt to identify independent predictors of rehabilitation outcomes. The study used a single, uniform method to measure rehabilitation success, MRFS-R, which was also used in our earlier studies (35, 42, 45).

The study has limitations. It is a retrospective study that used data from the computerized medical record with a sample of 144 participants. Several variables that might have affected the outcome of rehabilitation including the level of patient motivation and the extent of family participation in the rehabilitation process, were not collected. The follow-up period was only 1 year and there are no data on the patients’ functional levels on discharge from the department or in the long term.

In the present study the sample of the PF groups was compared to the sample of the group of patients with HF that were included in another study (35). Since this group of HF patients was limited to 138, we could not conduct matching between the groups in terms of age, sex, and comorbidity. This circumstance is, without doubt, a study limitation.

Finally, as noted above, due to potential selection bias in admission to the department, it may be difficult to generalize the results of this study to the general population of elderly patients with PF.

In summary, the present study found gaps between patients undergoing rehabilitation after HF and PF, with cognitive state the only predictor of a successful rehabilitation. In light of the increase in the rate of PF in the elderly population there is an urgent need for studies focusing on this condition in that population.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by Internal Review Board (Helsinki Committee) of the Soroka University Medical Center (Approval # SOR-0466-20; date of registration: 18/05/2021). The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required from the participants or the participants' legal guardians/next of kin in accordance with the national legislation and institutional requirements.

DK: Conceptualization, Data curation, Investigation, Writing – original draft. AG: Conceptualization, Data curation, Investigation, Writing – original draft. AA-A: Data curation, Investigation, Writing – original draft. EM: Data curation, Investigation, Writing – original draft. ES: Data curation, Investigation, Writing – original draft. EZ: Conceptualization, Data curation, Investigation, Writing – original draft. DS: Conceptualization, Data curation, Investigation, Writing – original draft. NV: Data curation, Investigation, Writing – original draft. TF: Conceptualization, Data curation, Formal analysis, Methodology, Writing – original draft. YP: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing.