Femoral neck fractures can have poor function if healed in a shortened position

SAN FRANCISCO – A multicenter study on patient function after having femoral neck fractures fixed with cancellous screws found that femoral neck shortening as well as varus collapse are common.

"Femoral neck shortening occurred in two out of three fractures regardless of whether they were displaced or nondisplaced, and it had a clearly negative impact on functional outcome," said Michael P. Zlowodzki, MD during the American Academy of Orthopaedic Surgeons 75th Annual Meeting, where he presented the results of a large multinational study.

Multinational study

Based on an initial pilot study, the investigators hypothesized that the incidence of neck shortening would be high. In order to avoid confounding the results by associated injuries, the study population consisted only of patients with isolated femoral neck fractures which healed neck after fixation with multiple cancellous screws. Patients were recruited at Regions Hospital, the University of Minnesota, St. Paul, Minn.; McMaster University, Hamilton, Ontario; Aarhus University, Aarhus, Denmark; and the University of Mississippi in Jackson.

Overall, 660 patients were screened based on their ICD-9 diagnostic code. Of those, 127 patients were eligible for inclusion and 70 patients consented to participate. The average age was 71 years and the average follow-up was 20 months.

Outcome measures included SF-36 scores and the EuroQual-5D. Investigators also assessed patients' pain, limping and use of assistive devices to ambulate.

Shortening assessment

Three independent physicians evaluated femoral neck shortening using recent follow-up radiographs, grading it as mild (0 mm to 5 mm), moderate (5 mm to 10 mm) or severe (>10 mm). They found approximate equal numbers of patients in each grade of shortening.

"Despite inequalities in the rotational profile between radiographs, surgeons' assessment of shortening was very reliable with an interclass correlation coefficient of 0.7, which is considered quite high," Zlowodzki said.

In a similar manner, investigators graded the amount of varus collapse in patients as mild (<5°), moderate (5° to 10°) or severe (>10°). Although Zlowodzki did not detail the collapse findings during his presentation, in the abstract it was noted that varus collapse occurred in 39% of patients, which correlated highly with shortening.

Functional outcomes

Patients with severe shortening of their hip had significantly lower SF-36 physical functioning scores – no/mild vs. severe shortening: 74 vs. 42 points. "Similar important effects occurred with moderate shortening suggesting a gradient effect – no/mild vs. moderate shortening: 74 vs. 53 points," Zlowodzki said. "Shortening also resulted in significantly lower EQ5D index scores. As expected, varus collapse also had a negative impact on patients' physical function; however, varus collapse was less common and it had a smaller impact on the assessed functional outcome parameters than shortening."

Preinjury, only one patient used an assistive walking device. Postoperatively, 27 patients used them: 15% in the no/mild shortening group, 43% in the moderate shortening group, and 83% in the severe shortening group.

"Similar results were found for the incidence of limping," he said.

What investigators learned

"The debate about the optimal treatment for femoral neck fractures is highly dogmatic, with many surgeons claiming that dynamic sliding of the fracture fragments along the implant during weight-bearing is vital for fracture healing," Zlowodzki said. "Yet, there is no clear poof for its necessity, and with this strategy, outcomes have not improved over the last 30 years even one bit.

"It is well known that compression stimulates fracture healing; however, it is not clear how much is needed or how much is too much," he added. "There is no evidence to suggest that dynamic fracture site compression is needed in addition to compression that can be achieved during the operative procedure, provided that the fracture can be reduced reasonably well. We believe that treatment options like multiple cancellous screws emphasize biology over mechanical stability, and that future studies will have to determine whether or not this is also true for sliding hip screws."

In addition, "We have not been critical enough with our results in the past. We considered a healed fracture a success and didn't pay enough attention to the patent's functional abilities. Keep in mind that the shortening issue is an additional problem on top of the actual fixation failure rate of up o 20% to 40% in certain fractures, which in itself is unacceptable," Zlowodzki said.

For more information:
Michael Pawel Zlowodzki, MD, can be reached at University of Minnesota, Department of Orthopaedic Surgery, 2512 S. 7th St., Room 200, Minneapolis, MN 55454; fax 651-254-1519; e-mail: zlowo001@umn.edu. He received research/institutional support from the Osteosynthesis and Trauma Care Foundation and AO North America.

Reference:
Zlowodzki MP, Brink O, Switzer JA, et al. Patient function following femoral neck shortening after screw fixation of femoral neck fractures. Paper #382. Presented at the American Academy of Orthopaedic Surgeons 75th Annual Meeting. March 5-9, 2008. San Francisco.