Spencer L. Butterfield

Acute mortality associated with injuries to the pelvic ring: the role of early patient mobilization and external fixation.

PURPOSE To analyze the effect on mortality of a protocol for early mobilization with external fixation of patients with pelvic ring injuries. METHODS From 1981 through 1988, 605 patients with pelvic ring fractures and dislocations were treated. In 1982, a protocol for early external fixation of hemodynamically unstable patients and those with structurally unstable pelvic fracture patterns to achieve early mobilization to an upright chest position was initiated. Mortality rates were compared between 1981 (pre-protocol), 1982 (transitional), and 1983 through 1988, after initiation of a protocol of care that included external fixation of the pelvic injury. No statistical changes occurred from 1983 through 1988. RESULTS Mortality rates in pelvic ring injury patients fell from 26% in 1981, to 6% in 1983 through 1988 (p < 0.001), whereas during the study period the mean injury Severity Score (ISS), 23, did not change. The mortality rate of a group of consecutive patients with comparable ISSs, but without pelvic ring injuries did not change. The mortality rate in patients with systolic blood pressure < 100 mm Hg at admission fell from 41% in 1981 to 21% 1983 through 1988 (p = 0.0001). Mortality in patients with closed head injuries associated with pelvic ring injuries fell from 43% in 1981 to 7% from 1983 through 1988 (p = 0.0001). The proportion of patients undergoing external fixation rose from 3% in 1981 to 31% in 1983 through 1988 (p = 0.0001). CONCLUSIONS An organized protocol including external fixation and early patient mobilization to an upright chest position reduced mortality associated with injuries of the pelvic ring. Orthopedic stabilization of major skeletal injuries should be viewed as part of patient resuscitation, not reconstruction.

Harvesting of autogenous cancellous bone graft from the proximal tibial metaphysis : a review of 230 cases

A study was undertaken to review the results of donor site morbidity of cancellous bone grafts obtained from the proximal tibial metaphysis. Two hundred six patients who underwent 230 proximal tibial bone graft harvestings were reviewed. Patients with lower-extremity fractures or nonunions who required cancellous bone grafts and would be non–weight-bearing for at least 6 weeks were selected to undergo the procedure. Minimum length of follow-up was 4 months, with an average length of follow up of 20.4 months. The proximal tibial metaphysis was found to supply an adequate amount of graft for all procedures involved, with a complication rate related to graft donor site of 1.3%. This compares favorably to a previously published report on bone grafts taken primarily from the illiac crest. The postoperative appearance of the proximal tibia may be permanently altered by the procedure, but weight-bearing after 6 weeks appears safe. The proximal tibial metaphysis is a useful site for obtaining cancellous bone graft and is associated with a low morbidity.

Complications of Seidel intramedullary nailing of narrow diameter humeral diaphyseal fractures.

Forty acute blunt fractures of the humeral diaphysis were treated with Seidel nails (9-mm diameter) between April 1988 and August 1992. Follow up was available for 36 patients. Average Injury Severity Score (ISS) was 22. Forty-eight pelvic and lower extremity fractures necessitated humeral weight bearing in 23 patients. Patients were grouped by canal diameter measured at the point of distal humeral fixation: < or = 9 mm or > or = 10 mm. Data were analyzed with regard to complications, fracture pattern, and time to union. Due to difficulties reaming the humeral canal, five different reaming systems were used. In the < or = 9 mm group (N = 12), there were seven complications (58%). Two patients had iatrogenic comminution distal to the nail. The two open fractures in this group (II, IIIA) developed wound infections, and one, osteomyelitis. Three patients had nonunions: one was lost to follow up at 6 months, and two united at 41 and 74 weeks after exchange nailing and bone grafting. Average time to union was 21 weeks. Six of the seven complications occurred in patients who required reaming of a long, tight segment of distal canal. In the > or = 10 mm group (N = 24), there was one complication of iatrogenic comminution (4%). There were four open fractures (three, grade II; one, IIIB) with no infections. All fractures united (average = 10 weeks). The differences in complications (P = .001) and union (P = .04) between groups were significant. Other complications were not associated with canal diameter or union. There were seven radiographic failures of the distal locking device (19%). Four nails (11%) were left prominent in the shoulder due to technical or equipment failures, and were eventually removed. Four patients (11%) had residual shoulder stiffness (three due to neurologic injury). These primary data suggest use of the Seidel nail is associated with a higher complication rate in humeri with canal diameters < or = 9 mm.

Vertical shear injuries: is there a relationship between residual displacement and functional outcome?

BACKGROUND Residual vertical displacement is often cited as being related to poor outcome in patients with pelvic injuries. This study attempts to clarify the relationship between residual vertical displacement and functional outcome. METHODS From 1982 to 1989, over 500 patients with pelvic ring injuries were treated at two Level I trauma centers. Thirty-three patients with vertical shear (Tile C) fractures and residual displacement (2-52 mm) were evaluated. Outcomes were quantified by using SF-36 Short-Form Health Survey (SF-36) and the Iowa Pelvic Score (IPS). RESULTS There was no correlation between IPS or SF-36 scales and residual vertical displacement. The IPS correlated (p<0.05) with seven of eight SF-36 categories, excluding mental health. Patients reporting limp and leg length discrepancy also correlated with the IPS and select SF-36. CONCLUSION Pelvic injuries showed no correlation between functional outcome and residual vertical displacement suggesting other factors. The degree of residual vertical displacement does not affect functional outcome.

Nonreamed nailing of tibial diaphyseal fractures in blunt polytrauma patients

Summary: The efficacy of nonreamed nailing as the treatment of choice of unstable blunt tibial diaphyseal fractures was studied. From March 1, 1990, through August 31, 1991, 72 patients with 74 fractures that required fixation were treated. One patient died and six were lost to follow-up, leaving 65 patients with 67 fractures. Follow-up averaged 21 months (range 5—43). Fishers exact and logistic regression analyses were used to compare grades of open fractures, comminution as classified by Winquist, and dynamic and static nailings. The failure rates of 51 titanium and 16 stainless steel nails were compared. Times to union were compared by the log rank statistic method. The average time to union was 32 weeks with 26 (39%) additional operations required to achieve union; 13 dynamizations (12 successful), 12 exchange nailings (11 successful), and one plate and bone graft. The rate of reconstructive procedures to achieve union was a more sensitive indicator of difficulties achieving union than was time to union. Reoperation rates were 33% for closed or grade I and II fractures compared with 46% for grade III fractures (NS). Among closed grade I and II static versus dynamic nailing, times to union were 36 versus 25 weeks (p < 0.01), and the reoperation rates were 44% versus 13% (p < 0.04). Winquist I and II fractures required a 24% reoperation rate versus 53% for grade III and IV and segmental fractures (p<0.01). Static locked fractures required a 48% reoperation rate versus 12% for dynamic locked fractures (p < 0.01). A logistic regression analysis demonstrated that locking mode was the most important factor in determining reoperation rates. Fifteen additional reoperations for infection, broken or painful implants, or to remodel bones that united with an incomplete circumference of cortex were performed. With an additional 12 elective nail removals, the total reoperations numbered 53 (79%). Titanium alloy nails had a 2% failure rate versus 25% for stainless steel nails (p < 0.01). Two of 28 (7%) grade III fractures became infected. All fractures united within 10° of normal alignment and 1 cm of length. Nine (13%) united with an incomplete cortical circumference, refractory to dynamization and full weight bearing. Thirteen of the 58 (22%) fractures available for an evaluation of ankle motion were symptomatic, with <10° of dorsiflexion. Nonreamed nails are an option in blunt polytrauma patients where immediate stability is critical and in grade III open and other high-energy fractures where secondary operations are acceptable. In other low-energy closed or grade I and II fractures (especially those statically locked), the use of nonreamed nails must be questioned

IMMEDIATE PLATE FIXATION OF HIGHLY COMMINUTED FEMORAL DIAPHYSEAL FRACTURES IN BLUNT POLYTRAUMA PATIENTS

From January 1982 through December 1988, 150 patients with 153 Winquist Class III and IV comminuted diaphyseal femur fractures due to high energy blunt trauma were treated with immediate plate fixation. A total of 260 major general surgical systems were injured in 150 patients. Forty-nine patients did not have adequate preoperative spine radiographs due to positioning or time factors. Nineteen patients had spine fractures; nine were diagnosed post-femoral fixation. The average injury severity score (ISS) was 22.7. Three patients died (2%). Our institution predicted mortality with this ISS for patients without pelvic or femur fractures at 15% (P = .0003). Six patients moved to other states and three were lost to follow up due to noncompliance. One of us reviewed 141 fractures in 138 patients at a minimum of 12 months follow up and completion of treatment. Forty-nine fractures were open; 8 grade I, 25 grade II, 10 grade IIIA, 4 IIIB, 2 IIIC. A total of 153 pelvic or ipsilateral major orthopedic injuries were present in 141 fractures. An additional 188 major associated orthopedic injuries were noted. The average time to union was 17.2 weeks. One plate was applied in 11 degrees of varus. Five plates failed from fatigue and five from repeat traumas. Seven plate failures were rodded and healed within 8 weeks. There was one persistent nonunion. One fracture, open IIIC, became infected after uniting. One patient has 110 degrees of knee motion and 140 fractures have greater than 130 degrees of knee motion. Plate fixation is a safe technique for immediate femoral stabilization in the face of high energy blunt trauma. Failures occur late and are easy to reconstruct. Intramedullary nails are the preferred method of reconstruction. Ultimate knee function is excellent. Infection rates (1/49) in open fractures are acceptably low.

Clandestine femoral neck fractures with ipsilateral diaphyseal fractures

Summary: We present a protocol for diagnosis of all femoral neck fractures associated with ipsilateral femoral diaphyseal fractures. A 30% incidence of delayed diagnosis has been reported by other investigators. Between 1982 and 1990, we have treated 32 patients with ipsilateral femoral neck and shaft fractures due to blunt trauma. Only 22 femoral neck fractures were diagnosed on prediaphyseal fixation radiographs. This left the 10 patients in this study who had normal prediaphyseal fixation radiographs and were subsequently found to have femoral neck fractures. The ipsilateral femoral neck fractures were found through a retrospective chart and radiographic review of all 555 femoral diaphyseal fractures identified through our trauma and fracture registries. The clinical and radiologic techniques for diagnosing the femoral neck fractures were presented. The time to union of the femoral shaft and neck was determined, and a preliminary radiologic assessment of the vascularity of the femoral head was made. Ten femoral neck fractures (31%) with normal preoperative radiographs were diagnosed after femoral diaphyseal fixation. One patient did not have a post-diaphyseal fixation radiograph. An incidental radiograph at 6 weeks showed a mildly displaced femoral neck fracture in an asymptomatic patient. At 16 weeks the patient became symptomatic, and a repeat radiograph showed the fracture. Five fractures were diagnosed in asymptomatic patients on routine post-femoral fixation radiographs. Two patients had normal post-femoral fixation radiographs, became symptomatic, and had their femoral neck fractures diagnosed on repeat radiographs at 3 and 7 days. One patient had normal pre- and postfixation radiographs, and on a 25-day routine femoral radiograph, the femoral neck fracture was diagnosed. In one patient, a femoral neck fracture was diagnosed during insertion of an intramedullary nail. Nine of the femoral diaphyseal fractures were plated because these were blunt polytrauma patients. All fractures were Pauwels classification III. Our study demonstrates that routine preoperative radiographs of the hip are inadequate to diagnose all femoral neck fractures associated with ipsilateral femoral diaphyseal fractures. We suggest (a) routine postoperative radiographs, (b) routine 2- to 4-week postinjury radiographs, and (c) radiographs any time a patient becomes symptomatic. Our incidence of delayed diagnosis of femoral neck fractures is similar to that reported in the literature. It is possible to have a mildly displaced femoral neck fracture and to be asymptomatic. Femoral neck fractures seen after intramedullary nailing are not due to improper technique.

THE ANTERIOR ACROMIAL APPROACH FOR ANTEGRADE INTRAMEDULLARY NAILING OF THE HUMERAL DIAPHYSIS

At three institutions, 71 humeral intramedullary nails were inserted into the shoulder; 67 were reviewed at 6 months and at completion of treatment. Fifty-one utilized the anterior acromial approach and 16 were inserted lateral to the acromion. Shoulder motion was rated as: excellent (asymptomatic and within 15 degrees of normal); good (normal daily function within normal motion); and poor. Nails were also inserted into the humeral diaphysis of eight cadaver shoulders. Fifty-one nails were inserted via the anterior acromial incision; 48 were graded as excellent, one as good, and two with traumatic axillary neuropathy and reflex sympathetic dystrophy as poor: Sixteen nails were inserted lateral to the acromion; 8 were rated, 7 good, and 1 poor. Motion returned in an average of 17 weeks (range:0-29). The greatest clinical concern is not ultimate shoulder function, but the rate of return. The authors conclude that either the anterior acromial approach or an extraarticular entry portal must be utilized for antegrade humeral diaphyseal nailing.

Comparison of reamed and nonreamed solid core nailing of the tibial diaphysis after external fixation: a preliminary report

Summary: From July 1982 to March 1990, 32 patients had an external fixator applied to treat a tibial diaphyseal fracture and subsequently underwent intramedullary nailing: 16 with reamed and 16 with nonreamed solid core nails. Indications for surgery were 12 atrophic and 1 hypertrophic nonunion in each group. The balance were either planned conversions or inadequate external fixators due to head injuries. All but one were seen by an author at a minimum of 1 year. Among the reamed nails, 3 fractures were grade III B. Two patients had pin tract infections, and there were no prenail wound infections. All infections were clinically inactive at the time of nail insertion. Postnail, 7 patients became infected, requiring 12 debridements and 2 procedures to achieve union. One patient had a plate applied 44 weeks postnail and was lost 48 weeks postnail with a persistent infected nonunion. The average time to union was 26 weeks. In the nonreamed solid core group, 2 fractures were grade III A and 5 grade III B. There were 2 pin and 5 prenail wound infections. One nail was inserted across an active pin tract infection. One tibia became infected postnail (p=.04). The fractures united at an average of 14 weeks postnail (p=.036). Two debridements to control infection but no further procedures to achieve union were necessary (p=.003). When tibial reconstructions following external fixation are required, nonreamed solid core nails are efficacious and may be preferable to reamed nails.

Ipsilateral femoral neck and shaft fractures: An overlooked association

A total of 304 patients with injuries to the femoral shaft and ipsilateral hip presented between 1984 and 1990. Some 253 of them suffered fractures of the femoral shaft and dislocated hips or fractures of the acetabulum, and 51 of these sustained fractures of the femoral shaft and neck or trochanteric region. Of this latter group, 20 patients had a combination of femoral shaft and neck fractures, and 31 had a combination of femoral shaft and trochanteric fractures. All of the trochanteric injuries were demonstrated on the initial radiographs. However, in 11 of the patients with combined femoral shaft and neck fractures, the diagnosis was delayed by as much as 4 weeks. This delay related to the fact that these fractures tended not to separate in the initial evaluation period and that there was external rotation of the proximal femoral fragment due to the femoral shaft fracture. Good preoperative and, in particular, good postoperative radiography of the hip is essential to make the diagnosis. Although orthopedic surgeons have been aware of this combination of injuries since 1953, radiologists have not.