Hip Fractures: Therapy, Timing, and Complication Spectrum

Abstract

Objective: Investigation of the treatment of femur fractures and the type of femur fracture-associated complications regarding timing of surgery and length of hospital stay.

Methods: In this retrospective cohort study, a total of 358 hip fractures were evaluated retrospectively from 1 January 2008 until 31 December 2010 at a level I trauma center in Germany. Inclusion criteria was age >18 years and a proximal femur fracture. Both sexes were evaluated. Mean age was 75.5 years, most patients were female (63.7%). Intervention was the operative treatment of proximal femur fracture. Outcome parameters were time until surgery, complications, reoperations, mortality, and length of hospital stay.

Results: Among the proximal femur fractures (n = 358), 46.6% were pertrochanteric, 11.2% subtrochanteric, and 42.2% femoral neck fractures. Operation upon hip fractures was managed regularly within 24 hours of injury (73%; mean for femoral neck: 28.3 hrs.; mean for pertrochanteric fractures: 21.4 hrs.; mean for subtrochanteric fractures: 19.5 hrs.). Delayed treatment, as well as implantation of hip total endoprosthesis (TEP), increased the overall length of hospital stay (15.4 vs 17.6 days; 18.1 vs 15.8 days). Accordingly, surgical procedures performed within 24 hours of injury resulted in a shorter hospital residence. Longest delay of operation was measured for hip fractures (28.3 hrs.). In 351 patients, secondary injuries were detected in 94 individuals (26%), with fractures being the most common secondary injury (n = 40). We recorded postoperative complications of nonsurgical and surgical origin, and 33.6% of our patient cohort displayed complications. Complications were distributed among 118 patients. There was no significant difference in complications regarding the time of operation, with most nonsurgical and surgical complications appearing within 24 hours after operation (n = 110 vs n = 31). Nonsurgical complications, such as anemia (n = 49) and electrolyte imbalances (n = 30), were observed more frequently than surgical complications (n = 107 vs n = 34); however, these complications were reduced by delay in surgery (82.0% in 6-24 hrs. vs 74.2% in ≥24 hrs.). Anticoagulant therapy and age did not affect postoperative complications. The hospital mortality of patients was 6.2%. Follow-up was restrained to ambulatory visits in the clinic.

Conclusions: Surgical management of hip fractures performed within 24 hours of injury minimizes hospital stay. We did not detect significant differences in the spectrum or number of complications regarding delay of surgery. Surgical complications mainly occur with rapid primary care, and medical complications can be reduced by more intensive preparation of patient and operation procedures.

Keywords: Age traumatology; Complications; Femoral neck fractures; Hip fractures; Pertrochanteric fracture.

Figure 1

An early operative treatment reduced the duration of the hospital stay. (A) In this cohort, pertrochanteric femur fracture was the most common injury, followed by femoral neck fracture. (B) In hip fracture, PFN was used most frequently (40%), followed by DHS (28%). (C) Mean length of hospital stay for THP was 18.1 days (compared to all other with a mean of 15.8 days). (D) Mean age of patients operated on within the first 6 hours of injury was 74.1 years and did not differ significantly from the other groups. An early operation led to a length of stay of 15.5 days. The group of patients operated on either within or after 24 hours of injury differs significantly in length of stay (t‐test, P = 0.0374).

Figure 2

THP‐related procedures are associated with the longest delay until surgery. (A) While DHS and PFN were implanted regularly within 24 hours of injury, THP and HHA were regularly implanted >24 hours after the accident (THP vs PFN: *; HHA vs PFN *, PFN vs Screw: **, PFN vs others: **, PFN vs no OP: **; two‐way ANOVA, Bonferroni post‐test). (B) THP was implanted after an average of 48 hours, HHA after an average of 34 hours. Operation with DHS or PFN was significantly faster compared to THP (P = 0.0001 [***] and 0.0009 [***] respectively, One‐way ANOVA, Tukey's post hoc test 5% level of significance).

Figure 3

Based on ASA status and mortality, age did not affect the clinical outcome. (A) ASA‐status did not differ significantly among the groups. Anticoagulants were administered to more patients the longer their operation was delayed, while in the >72 hours group, only 40% were treated. Immediate operation was significantly rarer when anticoagulants were taken (Chi‐Square test with 5% level of significance, P = 0.0152). (B) Over all age groups, operation was mostly performed within 24 hours of injury (P = 0.0058 [**], two‐way ANOVA with 5% level of significance). The deceased patients were distributed equally among the age groups and there were no differences in mortality between the groups (P = 0.7486, one‐way ANOVA with Tukey's post hoc test and 5% level of significance). (C) The deceased patients that were operated on within 24 hours of injury were slightly older than those operated on after >72 hours. In the latter group, the survivors were older than the deceased patients, although this finding was not significant (P = 0.65; t‐test, unpaired, 5%‐level of significance). (D) Operation on femoral neck fractures took place an average of 28.3 hours after the time of injury, although most patients were operated within 24 hours of injury. Pertrochanteric fractures were regularly operated on within 24 hours of injury, with a mean time until operation of 21.4 hours.

Figure 4

Hematoma and anemia were the most common complications. (A) Concomitant injuries occurred in 25.8% of femoral neck fractures, with 0.67 injuries on average per patient. Pertrochanteric fractures had 24.8% concomitant injuries, with 0.33 injuries per patient. Subtrochanteric fractures had the most concomitant injuries (46.5%), with 1.5 per patient on average. (B) Most complications were measured if the operation was performed within 6 to 24 hours of injury, while most surgical problems occurred within 6 hours of surgery. Age did not differ among the groups (P = 0.4195, one‐way ANOVA, Tukey's post hoc test, 5%‐level of significance).

Figure 5

Non‐Surgical and surgical complications show a predominance if operation was performed within the first 24 hours of injury. (A) The relative frequency of nonsurgical complications shows anemia (37.7%), electrolyte imbalances and transition syndrome as the most common complications, while (B) frequency of surgical complications demonstrates that hematoma/healing disturbances were most common (38.5%). (C) When operations were performed within 24 hours of injury, anemia and transition syndrome were most common in patients; however, anemia became rare with a longer period before operation (without leading to significant differences, one‐way ANOVA with Tukey's post hoc test). (D) In most patients, no surgical complications were measured. When operations were performed within 24 hours, mechanical complications and hematoma/healing disturbances were most likely to occur in patients (no significant differences in one‐way ANOVA with Tukey's post hoc test).