This study began in 2004 prior to the clinical trials registration requirement. After obtaining institutional review board approval, we conducted a prospective randomized study at a level-I trauma center between July 2004 and September 2007. One hundred and sixty-four patients with an intertrochanteric proximal femoral fracture were admitted to our hospital and were considered for the inclusion in the study.
The patient inclusion criteria were (1) an age of eighteen years or older and (2) an A1 or A2 AO/OTA intertrochanteric proximal femoral fracture. The PCCP is not indicated for femoral neck fractures, subtrochanteric proximal femoral fractures, and reverse obliquity intertrochanteric proximal femoral fractures. Exclusion criteria were (1) existing or previous fractures of the same or contralateral hip, (2) other fractures or injuries that could confound the outcome measures, (3) abnormalities that may confound the outcome measures of interest, (4) an inability to appear for follow-up for any reason for one year after surgery, and (5) an inability to understand the purpose of the study and/or an inability to give informed consent for inclusion. Altogether thirty-five patients were excluded on the basis of these criteria, and sixty-three declined to participate, leaving sixty-six included patients (Fig. 1).
Once the subject's eligibility was confirmed and consent was obtained, he or she was assigned to one of two treatment methods according to a computer-generated randomization schedule that was developed prior to beginning the study. The randomization order was computer-generated in a 1:1 ratio on the basis of balanced blocks of ten patients. The randomization order was well documented and known only by the trial coordinator. Sequentially numbered, sealed, opaque envelopes, which defined the treatment assignment, were opened as the subjects were enrolled. Of the sixty-six patients who consented to participate in the study, thirty-three were randomized to be treated with a PCCP and thirty-three, with a sliding hip screw.
Demographic data were collected, and the mobility of the patients before the fracture was determined. Intraoperatively, blood was collected in the pouch of the sterile barrier drape used during surgery. The amount of blood collected was measured by the surgeon and was confirmed by the anesthesiologist. The operating time was recorded as well as intraoperative transfusions, complications, and incision lengths. Surgery was done by seven attending surgeons and/or residents with direct supervision by an attending surgeon.
Surgical Technique: Gotfried PCCP
The PCCP system includes a plate with a chisel end for insertion through the vastus lateralis onto the lateral femoral cortex and along the femoral shaft, two telescoping compression neck screws, and three shaft screws for distal fixation (Fig. 2). The first step involves the reduction of the fracture. With use of the posterior reduction device, all posterior sag is eliminated. If possible, the neck-shaft angle is reduced to 135°. A small lateral incision is made at the level of the lesser trochanter under radiographic guidance. The plate, attached to an introducer (external jig), is passed through this incision. The second step is proper placement of the plate. Through a second, more distal, incision, a bone hook or clamp secures the plate to the femoral shaft. A template placed on the fluoroscopy screen is used to determine appropriate plate placement. Adjustments are made to ensure that the placement of the main guide (first drill) will pass close to the calcar on the anteroposterior fluoroscopic view and in the center of the femoral head and neck on the lateral view. It is preferable to position the neck screws parallel to the neck in both planes. The bone hook is secured tightly to ensure no movement of the plate (see Appendix). The main guide is drilled to within 5 mm of the articular surface (see Appendix). After the use of a series of sequential reamers that are placed through sequential sleeves in the introducer, the inferior neck screw is placed. The neck screw is captured in the sleeve, which locks into the plate at a 135° angle. Next, the three cortical shaft screws are placed with use of the introducer to ensure correct placement. Finally, the superior neck screw is placed after main guidewire drilling and sequential reaming (see Appendix). The wounds are then irrigated and closed with use of suture and skin staples. Skin incisions typically measure 25 mm in length.
Surgical Technique: Sliding Hip Screw
The sliding hip screw includes a plate with a variable number of shaft screws and a neck screw with an angle of 130°, 135°, or 140°. Standard technique was used to place the sliding hip screw implant in the center of the femoral head on the anteroposterior and lateral projections with use of fluoroscopic guidance.
Functional status was a patient-reported outcome recorded by the patients preoperatively and postoperatively. Radiographs were made preoperatively, during surgery, and at each follow-up visit. Assessment of outcomes and radiographic analysis was done by the senior author (E.Y.).
Following surgery, the patients were evaluated at two weeks, four weeks, and every month thereafter for one full year after surgery. At each visit, the clinical examination was performed with use of a dedicated questionnaire, functional recovery score, Short Form (SF)-36, and visual analog scale (VAS) for pain, and complications were noted. Postoperative level of function was scored on a scale of 1 to 4, with 1 being the best and 4 being the worst. The time to achieve this function was determined from the questionnaire and was measured in weeks.
All patients began weight-bearing as tolerated after surgery and were out of bed on the first or second postoperative day. Physical therapy was initiated within four days of the procedure, depending on a variety of factors.
Data were entered into a Microsoft Access database (Microsoft, Redmond, California). Descriptive data were summarized as the mean and the standard deviation, while frequency counts and percentages were used for the dichotomous data. Continuous data were compared with use of t tests, while chi-square analyses were used for dichotomous data. Significance was assessed at a p value of 0.05. SAS software (version 9.1; SAS Institute, Cary, North Carolina) was used for all analyses.
Source of Funding
There was no external funding for this study.
A table in the Appendix summarizes the demographic data. Nineteen men and forty-seven women were enrolled in the study, and the mean age was seventy-seven years. The men had a mean age of sixty-nine years (range, thirty to ninety-three years), and the women had a mean age of eighty-two years (range, forty-eight to 101 years). Eleven men and twenty-two women were randomized to receive the PCCP, and they had a mean age of seventy-six years (range, thirty to 100 years). Eight men and twenty-five women were randomized to receive the sliding hip screw, and they had a mean age of seventy-seven years (range, forty-four to 101 years). The treatment groups were similar with respect to study variables (p > 0.05).
Mortality in the twelve months after surgery in the sliding hip screw group was nearly twice that in the PCCP group (27.3% vs. 15.2%). While it may be clinically important, this difference was not significant (p = 0.22).
Surgical data are summarized in a table in the Appendix. The mean surgical time was forty-eight minutes (range, twenty-two to 148 minutes) for the PCCP group and seventy-eight minutes (range, thirty-five to 137 minutes) for the sliding hip screw group. The mean combined incision length was 56 mm (range, 22 to 100 mm) for the PCCP group and 82 mm (range, 40 to 150 mm) for the sliding hip screw group. The mean blood loss was 41 ± 40.9 mL for the PCCP group and 101 ± 89.6 mL for the sliding hip screw group. Transfusion requirements varied between the treatment groups (0.72 unit for the PCCP group and 1.06 units for the sliding hip screw group), but the difference was not significant (p = 0.22). The perioperative transfusion rates are listed in a table in the Appendix. Operative time, incision length, and blood loss were all significantly different (p < 0.005) and favored the PCCP group.
More patients in the PCCP group returned to their original living environment (see Appendix), and fewer patients in that group (40% vs. 59%) required the assistance of walking aids, but neither difference was significant. The median follow-up period for surviving patients in both groups was thirty-six months.
Thirty-eight patients had functional recovery scores available for both the preoperative and the twelve-month follow-up evaluation. For the eighteen patients in the sliding hip screw group, the mean loss in functional status was 15.8 points (range, −86 to 12 points), while the twenty patients in the PCCP group had a mean loss of 10.7 points (range, −48 to 28 points) (p = 0.47).
The SF-36 scores at twelve months showed no significant difference in the physical function between the PCCP group and the sliding hip screw group (47.5 and 38.6, respectively). There was an advantage in general health perception, social functioning, and role emotional favoring the PCCP group. Pain at rest was slightly better at three and six months, and pain with activity was lower from three to twelve months. A comparison of VAS scores at rest and with activity is depicted in Figures 3-A and 3-B. The only significant difference in VAS scores occurred with activity at three months.
Any advantages associated with the PCCP in the treatment of AO/OTA A1 and A2 intertrochanteric proximal femoral fractures occurred early in the recovery period. These results included a shorter operative time, a smaller incision, less blood loss, and decreased need for transfusion while maintaining at least equivalent functional results.
To achieve an adequate reduction with use of the PCCP technique, a posterior reduction device plays an integral role by assisting the reduction and preventing posterior sagging of the fracture fragments. The use of this device may also reduce total surgery time9,10. Criticism of the surgical technique suggests that the PCCP is more demanding than the sliding hip screw9,11. In our experience, the learning curve was perceived to be only a few cases before the overall reduction of surgical time is achieved. Resident surgeons were involved in all of the cases. We started using the PCCP four and a half years before the start of this study12. In our initial series, no patient required conversion to an open procedure to insert this plate, and surgical time averaged sixty minutes.
Minimally invasive techniques are thought to improve surgical outcomes by reducing perioperative blood loss, soft-tissue damage, postoperative pain, and morbidity13. Elderly patients with associated comorbidities may avoid the hazards of blood transfusion, long anesthesia times, major tissue trauma, and slower return to function with the PCCP technique. Our results have been verified in recent prospective studies comparing the sliding hip screw and PCCP devices9-11.
In the late 1980s, biomechanical studies led to the recommendation for fixation of intertrochanteric proximal femoral fractures with an anatomical reduction14. Both the PCCP and sliding hip screw require anatomical and nearly anatomical reduction; however, unlike the sliding hip screw, the PCCP protects the lateral femoral wall from iatrogenic compromise15. Compression, impaction, collapse, and sliding are terms that have been used interchangeably when describing the effect of weight-bearing on the intertrochanteric fracture fixed with a sliding hip screw. These terms have recently been better defined16. Compression and impaction resulting from sliding of the screw within its barrel requires an intact lateral cortex and is desirable. The popularity of the cephalomedullary nail may be due to its ability to usually maintain a reduction despite the need for an intact lateral wall. Healing can often be achieved without compression at the fracture site.
Although there were no significant differences in the complications between the two implants in our study, collapse remains a considerable problem with the use of a sliding hip screw17. Bendo et al.18 reported that excessive collapse after treatment of unstable intertrochanteric fractures is equivalent to fixation failure and leads to leg-length discrepancy, abductor weakness, and gait abnormalities. Lateral wall failure has been almost eliminated with the use of the PCCP15. In the present study, there was no lateral wall fracture with the use of the PCCP, but we did not evaluate the sliding hip screw group for lateral wall fractures.
In a recent study on potentially unstable intertrochanteric fractures, Im et al.19 reported that iatrogenic comminution of the lateral cortex with use of a sliding hip screw was the most significant factor associated with loss of initial reduction, which led to poorer mobility scores despite complete fracture-healing. We believe that iatrogenic fracture of the lateral wall is associated with drilling of the lateral wall with the triple reamer used in the sliding hip screw technique. Large-diameter defects with a transverse orientation increase the risk of fracture20. Finally, Palm et al. reported that reoperation for an intertrochanteric hip fracture is associated with the occurrence of lateral wall fracture21.
The PCCP provides rotational stability through the use of double axis fixation in the femoral head fragment. When the device is used, the lateral wall is protected by incremental drilling of two small-diameter holes in the lateral cortex compared with the 16 to 32-mm drilling required for the screw barrel of the sliding hip screw. The PCCP also provides dual axis fixation in the proximal fragment, thus resulting in improved rotational control22.
In our study, the difference in the functional results between the two groups was not significant. The SF-36 and VAS scales favored the PCCP, especially at the three-month period. Janzing et al. also reported less postoperative pain with the use of the PCCP10. We postulate that the trend toward improved function and less pain is due to fracture compression against an intact lateral wall on postoperative weight-bearing in the PCCP group. This sort of controlled compression is less predictable with the use of the sliding hip screw because of altered integrity of the lateral cortex.
There are some limitations in our study. The most important is the low recruitment rate, which did not allow demonstration of significant differences in functional outcomes between these two surgical procedures. Sixty-three (38%) of 164 patients declined to participate, which is higher than that reported by Peyser et al.23, with thirty-nine (17%) of 225 patients declining to participate. Our patients wanted to know and decide the treatment before surgery.
Another limitation concerns the preexisting conditions and postoperative living arrangements that made it difficult for patients to keep follow-up appointments for the twelve-month period of the study, leading to the required use of telephone interviews to complete of the SF-36 and the functional recovery score questionnaires. Furthermore, with the growing popularity of intramedullary fixation for the treatment of intertrochanteric fractures, future studies may require a third study arm of patients treated with a cephalomedullary device.
In conclusion, the Gotfried PCCP system provides a minimally invasive approach for the treatment of intertrochanteric proximal femoral fractures that compares favorably with the surgical treatment of these fractures with use of the sliding hip screw. Compared with the sliding hip screw, the PCCP resulted in a shorter operative time, a smaller total incision, and decreased blood loss, while maintaining at least equivalent functional results.
NOTE: The authors thank Veronica Emma for coordinating the trial and for conducting follow-up assessments; Richard Ghillani and Azriel Benaroya for recruiting their patients into this study; Yuriy Akilova for conducting follow-up assessments; and Toni Lewis for help with the final submission.