This Article Extract Full Text (PDF) CME: Take the exams for this article: Trauma Test 20: Winter 2008 (publication date February 15, 2008; expiration... CME 4: October, November, December 2007 (publication date January 4, 2008; ... [Supplementary Material] Letters to the Editor: Submit a response Alert me when this article is cited Alert me when Letters to the Editor are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Reprints and Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Cole, P. A. Articles by Bhandari, M. Search for Related Content PubMed PubMed Citation Articles by Cole, P. A. Articles by Bhandari, M. Related Collections Specialty Update Adult Trauma Social Bookmarking             What's this?

What's New in Orthopaedic Trauma

¹ Department of Orthopaedic Surgery, Regions Hospital, 640 Jackson Street, St. Paul, MN 55101. E-mail address: peter.a.cole{at}healthpartners.com
² Department of Orthopaedic Surgery, University of California San Francisco, San Francisco General Hospital, 1001 Potrero Avenue, Room 3A36, San Francisco, CA 94110
³ Orthopaedic Trauma Service, Hamilton General Hospital, 237 Barton Street East, 6 North Trauma, Hamilton, ON L8L 2X2, Canada

Specialty Update has been developed in collaboration with the Council of Musculoskeletal Specialty Societies (COMSS) of the American Academy of Orthopaedic Surgeons.

Disclosure: In support of their research for or preparation of this work, one or more of the authors received, in any one year, outside funding or grants in excess of $10,000 from National Institutes of Health; National Institute of Musculoskeletal and Skin Disease (R01AR053645). In addition, one or more of the authors or a member of his or her immediate family received, in any one year, payments or other benefits of less than $10,000 or a commitment or agreement to provide such benefits from a commercial entity (small speaking honoraria related to AO or Zimmer trauma courses). Also, a commercial entity (Stryker Biotech) paid or directed in any one year, or agreed to pay or direct, benefits in excess of $10,000 to a research fund, foundation, division, center, clinical practice, or other charitable or nonprofit organization with which one or more of the authors, or a member of his or her immediate family, is affiliated or associated.

	Introduction

Top Introduction General Topics Clinical Axial Skeleton Appendicular Skeleton Appendix References

 
This Specialty Update in orthopaedic trauma will emphasize the most relevant clinical literature that was published between June 2006 and May 2007. Given the scope of the literature and the broad overview of trauma, we have focused our search to include journals that we believe to be most relevant to orthopaedic trauma. Our approach certainly is not exhaustive, however. For example, we did not translate non-English-language articles, nor did we cover hand and pediatric trauma, given that these topics will be reviewed in their specific subspecialty sections. Specifically, we searched the Cochrane Database, seven orthopaedic journals, and four high-impact medical journals (see Appendix). This search identified 8469 articles, of which 340 were selected for possible relevance. All of these abstracts were then reviewed by one of us (P.A.C.). After the abstract review, sixty-eight studies were chosen for this update, including nineteen Level-I studies, fifteen Level-II studies, seven Level-III studies, and twenty-seven Level-IV studies. A table in the Appendix summarizes each of the studies and classifies them according to level of evidence and study type. Of the sixty-eight clinical articles, forty-seven represented studies of therapy, sixteen involved prognosis, and five evaluated diagnostic tests.

This year, for the first time, we are pleased to complement this update with new material in the area of basic science. Under the guidance of one of us (T.M. III), a process has been established to hone in on important basic-science works relevant to the field, with the goal of elucidating information that we believe may be equally important to the clinical papers typically reviewed in this update. This process is not straightforward, however, as there is no level-of-evidence strategy to help to evaluate and contextualize basic-science studies. Although many of the techniques and strategies used in basic-science investigation are foreign to most clinicians, it has become increasingly clear that clinical and basic-science material often intersect and provide a great deal of relevance to the orthopaedic clinician.

This basic-science material will be presented under the heading of General Topics. Given the wide range of possible research fields that relate to musculoskeletal trauma and repair, the task of selecting a handful of published projects seemed daunting. However, fifteen recognized experts were polled and were asked to identify the one or two most important trauma-related research articles in their field. We then selected (perhaps arbitrarily) six articles for review that address vital areas of cutting-edge research.

We trust that you will be enriched by this year's orthopaedic trauma survey and hope that this overview will prompt you, and make it easier for you, to dive deeper into selections and topics that otherwise may be much more difficult.

	General Topics

Top Introduction General Topics Clinical Axial Skeleton Appendicular Skeleton Appendix References

 
Basic Science
Growth Factors
Musculoskeletal repair occurs through a series of events mediated by growth factors. Bone morphogenetic proteins (BMPs) have been identified as key factors that influence a variety of processes, including chemotaxis, cellular proliferation and differentiation, new blood-vessel formation, chondrogenesis, and osteogenesis.

The degree to which BMP-2 is required for fracture-healing was evaluated by Tsuji et al.¹. Those investigators created transgenic mice lacking limb-specific expression of Bmp-2 using a Prx1cre enhancer (Bmp2^c/c; Prx1::cre), thereby avoiding the embryonic lethality from complete loss of Bmp-2. While these mice did not appear to have defects in skeletal patterning, they did have development of dose-dependent deficits in bone mineral density relative to their heterozygous and control littermates. When unilateral femoral fractures were created in the mice lacking limb-specific BMP-2, the fractures failed to heal by Day 20 compared with the heterozygous and the control littermates, which healed by that time-frame. These differences manifested as delayed periosteal activation as well as the absence of a bridging callus and undifferentiated mesenchymal progenitor cells. Additionally, Bmp-4 and Bmp-7 transcripts were present at comparable levels in mice with and without BMP-2, suggesting that these BMPs cannot compensate for the absence of BMP-2 in fracture-healing. The authors concluded that BMP-2 is required for the initiation of fracture repair.

Developmental Biology
Fracture repair occurs through intramembranous and endochondral ossification. As many of the processes that govern bone development and repair are shared, investigators have gained insight into adult fracture-healing through their understanding of skeletogenesis. Chondrocyte hypertrophy is a key process in bone formation as cartilage matures and then is replaced by bone. Several factors, such as members of the Runt family of transcription factors (Runx2, Runx3), have been shown to be required for chondrocyte hypertrophy and osteoblast differentiation, but this past year, Nishio et al. demonstrated that Osterix, a transcription factor that acts downstream of Runx2, is also required for osteoblast differentiation and bone formation as well².

Arnold et al. further reported on the previously undescribed essential role of myocyte enhancer factor 2C (MEF2C) in the process of chondrocyte hypertrophy and bone development³. Although MEF2C is known to regulate muscle and cardiovascular development, transgenic mice that have a genetic deletion of MEF2C in endochondral cartilage have impaired chondrocyte hypertrophy, angiogenesis, ossification, and longitudinal bone growth. Furthermore, effects in bone deficiency of the MEF2C mutant mice can be rescued by a mutation in its co-repressor histone deacetylase 4 (HDAC4). The authors concluded that MEF2C controls bone development by activating the gene program for chondrocyte hypertrophy, which shares mechanistic commonalities with muscle and cardiovascular development.

The mechanical environment influences musculoskeletal development and repair, including mesenchymal differentiation into cartilage or bone. However, the mechanisms through which mechanotransduction occurs are poorly understood. Cilia are microtubule-based organelles present on the majority of mammalian cells that have various functions, including sensory perception and mechanosensation. Cilia formation requires intraflagellar transport (IFT), and mutations in intraflagellar transport result in the loss of cilia. Haycraft et al. generated a new conditional allele of the Ift88 gene and, using the Cre-lox system, disrupted cilia on different cell populations in the developing limb bud⁴. Disruption of the intraflagellar transport in the mesenchyme in the conditional mutant mice (prx1cre;Ift88) demonstrated multiple defects, including disorganization of the perichondrium and abnormal chondrocyte differentiation, as well as endochondral bone formation. Given that intraflagellar transport results in impaired sonic hedgehog (SHH) signal transduction, and that indian hedgehog (IHH) regulates chondrocyte proliferation, the investigators evaluated IHH signaling in the prx1cre;Ift88 mutants and found it to be disrupted. In addition, the endochondral bone phenotype in these mice had similarities with other mouse models, suggesting multiple roles for intraflagellar transport and cilia in endochondral bone formation. Future studies involving this type of technology will shed additional light on the relationship between mechanical stimuli and the cellular and molecular response.

Tissue Engineering and Nanotechnology
Although bone tissue has an intrinsic ability to repair itself with new bone, critically sized defects require bone restorative procedures. Successful approaches will require advances in mechanically and biologically compatible scaffolds, inductive factors, and progenitor cell biology. Using a novel dynamic oscillating culture technique, Weinand et al. studied the possibility of developing bone tissue in vitro⁵. A three-dimensionally printed porous scaffold made of -tricalcium-phosphate (-TCP) was suspended in different hydrogels (collagen I, fibrin, alginate, and pluronic F127) and bone-marrow-derived differentiated mesenchymal stem cells. Histological analyses were performed at one, two, four, and six weeks, and radiographic, gene transcription, and biomechanical analyses were performed at six weeks. The combination of the scaffold, cells, and collagen I samples was the superior combination, with increased histological bone-tissue formation, radiographic opacities, expression of bone-specific genes, and improved mechanical strength under dynamic oscillating conditions. The authors concluded that in vitro bone tissue can be successfully formed with the proper combination of hydrogel (collagen I), scaffold (-TCP), progenitor cell, and mechanical stimulus.

One of the goals in developing biomaterials for use in tissue engineering and integration is to design scaffolds made of the proper material and configuration for adequate cell adhesion and biomechanical support. Nanotechnological approaches have been used to create materials that improve bone growth into materials. Popat et al. created a novel material with titania interfaces fabricated with controlled nanoarchitecture with use of anodization⁶. They studied the in vitro osseointegration of rat marrow stromal cells on the titania surfaces (nanotubular and flat) at different time-points. The cells cultured on the nanotubular surfaces showed greater cell adhesion and proliferation, alkaline phosphatase activity, and bone matrix deposition compared with those grown on the flat titanium surfaces. These different implanting surfaces were then implanted subcutaneously in rats and were evaluated histologically for an inflammatory response. No fibrosis or inflammation was seen surrounding the implants; therefore, the authors surmised that materials with nanotubular surfaces promote osseointegration, without an inflammatory reaction. Nanotopography is a highly promising field for improved cell-material interactions.

	Clinical

Top Introduction General Topics Clinical Axial Skeleton Appendicular Skeleton Appendix References

 
Growth Factors
This section reviews three studies involving the application of a growth factor on a bovine collagen sponge to a site of tibial pathology. Two studies involved bone morphogenetic protein-2 (BMP-2) (Infuse; Medtronic Sofamor Danek, Memphis, Tennessee), and one involved bone morphogenetic protein-7 (BMP-7) (Osigraft; Stryker Biotech, Limerick, Ireland). To date, the gold standard for the treatment of long-bone defects and nonunions has been autogenous bone-grafting. Donor-site morbidity and the limited supply of suitable bone graft make this solution problematic in some cases. In a randomized controlled study of thirty adult patients who had an open diaphyseal tibial fracture with a bone defect, autogenous bone graft was compared with allograft combined with recombinant human BMP-2 (rhBMP-2) on a collagen sponge (1.5 mg/mL)⁷. Fifteen patients were enrolled in each group. The mean length of the defect was not significantly different between the groups, and other variables such as smoking, age, comorbidities, and fracture type were equivalent between the groups. Ten patients in the autograft group and thirteen patients in the rhBMP-2/allograft group had healing (as determined on the basis of a blinded radiographic assessment) without further intervention, although an insufficient volume of autograft had been harvested in the cases of three patients in the autograft group. The mean estimated blood loss was significantly less in the rhBMP-2/allograft group. There was no significant difference in functional outcome according to the Short Musculoskeletal Function Assessment score. Only one patient had development of transient antibodies to bovine type-I collagen. The authors of the study concluded that rhBMP-2/allograft is safe and is as effective as traditional autogenous bone-grafting for the treatment of tibial fractures associated with diaphyseal bone loss.

A subgroup analysis of the combined data from two prospective clinical studies on the use of rhBMP-2 for the treatment of open tibial shaft fractures formed the basis for the second investigation⁸. Five hundred and ten patients with open tibial fractures were prospectively randomized to receive the control treatment (intramedullary nail fixation and routine soft-tissue management) or the control treatment combined with the use of an absorbable collagen sponge impregnated with one of two concentrations of rhBMP-2, which was placed over the fracture at the time of definitive wound closure. Only the control treatment and the Food and Drug Administration-approved concentration of rhBMP-2 (1.50 mg/mL) were compared in the subgroup analysis. Fifty-nine trauma centers in twelve countries participated, and patients were followed for twelve months postoperatively. Analysis of the subgroup of 131 patients with a Gustilo-Anderson type-IIIA or IIIB open tibial fracture revealed a lower rate of bone-grafting procedures (2% compared with 20%; p = 0.0005), a lower rate of invasive secondary interventions (9% compared with 28%; p = 0.0065), and a lower rate of infection (21% compared with 40%; p = 0.0234) in the rhBMP-2 group relative to the control group. On the other hand, analysis of the subgroup of 113 patients who were managed with reamed intramedullary nailing revealed no significant difference between the rhBMP-2 group and the control group. This population of patients was less severely injured, with a higher percentage of Gustilo and Anderson open type-I and type-II fractures, and the authors pointed out that this group was not sufficiently powered to show significant differences, although a statistical trend favored the rhBMP-2 group. Overall, the data in these two studies favored the use of rhBMP-2 for the treatment of severe open tibial fractures.

The third growth-factor study investigated a more heterogenous group of patients with distal tibial fractures (including open and closed fractures as well as intra-articular and extra-articular fractures) that were treated with a ring hybrid external fixator⁹. Twenty patients with distal tibial fractures that had been treated with external fixation as well as growth factor (rhBMP-7)-soaked bovine collagen were compared with twenty matched patients who had received similar treatment without rhBMP-7. Significantly more fractures had healed by sixteen (p = 0.039) and twenty weeks (p = 0.022) in the BMP group as compared with the matched group, despite a higher percentage of smokers, high-energy-fracture variants, and larger bone defects in the former group; the evaluations were not blinded. The mean time to union (p = 0.002), the duration of absence from work (p = 0.018), and the time for which external fixation was required (p = 0.037) were significantly shorter in the BMP group than in the matched group. Furthermore, the BMP group had fewer secondary interventions, leading the authors to conclude that rhBMP-7 can enhance the union of distal tibial fractures treated with external fixation.

Geriatric Orthopaedics—Risk Factors
It is difficult to sort out the relative effects of different risk factors for fragility fractures in the elderly as there are so many variables at work in this population. More large-scale clinical trials are being conducted to answer these questions. Interestingly, this year, additional insight has been gained regarding frequently hypothesized fracture risk factors such as a previous fracture or dementia, but new risk factors have emerged as well.

In the British Medical Journal earlier this year, a metaanalysis was published to present the evidence for strategies to prevent fragility fractures in residents of care homes and hospitals and to investigate the risk factors of dementia and cognitive impairment¹⁰. Of the 1207 references that were identified, forty-three studies (including sixteen randomized controlled trials) met inclusion criteria based on quality grading. When multifaceted interventions were employed in hospitals, the authors reported a rate ratio of 0.82 (95% confidence interval, 0.68 to 0.997) for falls but observed no significant effect on the number of fallers or fractures. When hip protectors were employed in care homes, the rate ratio for hip fractures was 0.67 (95% confidence interval, 0.46 to 0.98), but there was no significant effect on falls and there were not enough data on fallers. The authors concluded that there needs to be more research on the effectiveness of different single interventions in hospitals and care homes in order to understand which combination of interventions would have the greatest positive effect.

An Australian prospective study that was published in the Journal of the American Medical Association answered the broader question of refracture risk in the population at large¹¹. Although the study population consisted of an all-white cohort (based on regional demographics), this longitudinal study (from 1989 to 2005) investigated community dwellers, including 2245 women and 1760 men over the age of sixty years, who were followed for a median of sixteen years. Of the 905 women and 337 men with an initial fracture, 253 women and seventy-one men experienced another fragility fracture, making the relative risk of refracture 1.95 (95% confidence interval, 1.70 to 2.25) in women and 3.47 (95% confidence interval, 2.68 to 4.48) in men. The estimated rate of loss to follow-up was only 5.3%. Additionally, the absolute risk of subsequent fracture was similar for men and women, and this increased risk occurred for virtually all age-groups and for all clinical fractures and persisted for as long as ten years. Approximately 50% had a refracture by ten years, but the majority of those fractures occurred within the first five years. Femoral neck bone mineral density, age, and smoking were the best predictors of subsequent fracture in women, and femoral neck bone mineral density, physical activity, and calcium intake were predictors in men.

The optimal duration of treatment with alendronate for women with postmenopausal osteoporosis is unknown. A randomized, double-blind trial was conducted at ten clinical centers in the United States¹². One thousand ninety-nine postmenopausal women who had been managed with alendronate for the treatment of low femoral neck bone mineral density (<0.68 g/cm³) and other high risk factors for five years were randomized to treatment with alendronate at a dosage of 5 mg/d (n = 329) or 10 mg/d (n = 333) or with placebo (n = 437) for five more years (from 1998 to 2003). Women who discontinued the drug for five years showed a moderate decline in bone mineral density and a gradual rise in biochemical markers for bone turnover, but no higher fracture risk other than for clinical (not morphometric) vertebral fractures, in comparison with those who continued to receive alendronate. The authors concluded that discontinuation of alendronate for up to five years (after treatment for five years) should be considered except for women with a very high risk of clinical vertebral fractures.

It has been postulated that proton pump inhibitors that are used to suppress stomach acid production may interfere with calcium absorption and thus inhibit proper osteoclastic function, which could in turn affect fracture risk. One study investigated a cohort consisting of users of proton pump inhibitor therapy and nonusers of acid-suppression drugs who were older than fifty years of age¹³. In that study, 13,556 patients with a hip fracture and 135,386 controls were culled from a large database of general practices in the United Kingdom. The adjusted odds ratio for hip fracture associated with more than one year of proton pump inhibitor therapy was 1.44 (95% confidence interval, 1.30 to 1.59), and interestingly, the strength of the association increased with increasing duration of proton pump inhibitor therapy (adjusted odds ratio, 1.22 [95% confidence interval, 1.15 to 1.30] for one year; 1.41 [95% confidence interval, 1.28 to 1.56] for two years; 1.54 [95% confidence interval, 1.37 to 1.73] for three years; and 1.59 [95% confidence interval, 1.39 to 1.80] for four years). The data were analyzed while controlling for a multitude of possible confounding variables, demonstrating that proton pump inhibitor therapy is strongly associated with an increased risk of hip fracture.

Because it is well known that malnutrition is common among the elderly, the authors of a meta-analysis published by the Cochrane Collaboration sought to answer the question of whether nutrition supplementation protocols affect mortality after fracture¹⁴. Twenty-one randomized trials involving 1727 participants were included, although the authors acknowledged the relatively poor quality ratings of the trials. Eight trials evaluated oral multinutrient feeding, four trials examined nasogastric multinutrient feeding, four trials tested increased protein intake in an oral feed, and two trials tested intravenous administration of vitamin B1 and other water-soluble vitamins (one trial) or 1-alpha-hydroxycholecalciferol (one trial), with all of the trials showing no statistical evidence for decreased mortality. There was some weak evidence to suggest that protein supplementation may have reduced the number of long-term medical complications. One trial evaluating the role of dietetic assistants to help with feeding showed a trend toward a reduction in mortality (relative risk, 0.57; 99% confidence interval, 0.29 to 1.11). The conclusion of the review was that adequately sized trials are required to overcome methodological defects of the reviewed studies, and perhaps special attention should be placed on the role of dietetic assistants.

Beyond the broad systematic review on the effects of nutrition supplementation on post-fracture mortality, one recently published prospective randomized trial on the effect of a dietary intervention on postoperative morbidity among patients with hip fractures may give us cause for hope with dietary action¹⁵. In a nicely controlled study of eighty patients with intracapsular or trochanteric hip fractures, including only communicative, otherwise healthy, consentable individuals with no cognitive defects or comorbidities, the effect of a nutrition program was evaluated with regard to the morbidity outcomes. The patients were randomized to either the control group (comprising forty patients who received ordinary hospital food and beverage) or the intervention group (comprising forty patients who received a 1000-kcal daily intravenous supplement for three days, followed by a 400-kcal oral nutritional supplement for seven days). The results of the study revealed that the control group received only 54% and 64% of optimal energy and fluid intake, respectively, whereas the intervention group had close to optimal nutritional intake (p < 0.001 and p < 0.0001). The risk of fracture-related complications was greater in the control group than in the intervention group (70% compared with 15%; p < 0.0001). In addition, four patients in the control group, and none in the intervention group, died within 120 days postoperatively (p = 0.04), proving the efficacy of this nutritional program when used for healthy patients with a hip fracture.

The Mangled Extremity
The Lower Extremity Assessment Project (LEAP) study and its scientific offspring continue to pay dividends. The purpose of a study this year by the Evidence-Based Orthopaedic Trauma Working Group was to perform a systematic review of the literature in an effort to aid clinicians in decision-making regarding limb salvage as opposed to primary amputation for the mangled extremity¹⁶. Nine observational studies formed the basis of the review. While the length of hospital stay was similar for limb salvage and primary amputation, the length of rehabilitation and total costs were higher for limb salvage. Furthermore, patients managed with limb salvage required more procedures and were more likely to undergo rehospitalization.

Long-term functional outcomes (as long as seven years after the injury) were equivalent between the limb salvage and primary amputation groups, and the rates of self-reported disability and pain were similar as well. Functional outcomes worsened over time in both groups, and only about half of the patients in both groups returned to competitive employment two years after the injury. Clinical and psychosocial factors, including smoking, nonwhite race, uninsured status, having less than a high-school education, an income status below the poverty line, and low self-efficacy were all associated with poorer functional outcomes. Additionally, whereas it is common at the time of injury for patients to prefer limb salvage, the majority of patients with a failed limb salvage stated that they would opt for early amputation if they could decide again. These findings certainly indicate that strict patient selection judgment should be exercised in determining appropriate candidates for limb salvage.

	Axial Skeleton

Top Introduction General Topics Clinical Axial Skeleton Appendicular Skeleton Appendix References

 
Spine
The first two features in this section on the spine involved diagnostic studies, both of which attempted to provide a guide for prognosis and the need for surgery. The first, a retrospective study from the University of Massachusetts involving twenty-six consecutive unilateral facet fractures that did not initially meet surgical criteria (mechanical instability or unstable neurological deficit), was performed to determine computed tomographic features that lead to the failure of conservative treatment¹⁷. When patients who were successfully managed nonoperatively were compared with those who required operative decompression and stabilization, those with unilateral cervical facet fractures involving >40% of the absolute height of the intact lateral mass or an absolute height of >1 cm were at increased risk for the failure of nonoperative treatment (p = 0.0002 and p = 0.008, respectively), whereas no patient with a fracture involving <40% of the height of the lateral mass or an absolute height of <1 cm had failure of nonoperative treatment. The limitations of the study were its small sample size and the lack of a clear protocol for surgical indications; however, it may serve as a clinical guide and as a good basis for a prospective study.

Although little correlation has been shown between the degree of collapse of a vertebral body and the level of pain in patients with fragility fractures, previous studies have only been based on supine or standing lateral thoracolumbar radiographs. Toyone et al. studied 100 consecutively managed patients (median age, seventy-five years; range, sixty to eighty-nine years) who had back pain after a lower thoracic or upper wedge compression fracture resulting from low-energy trauma¹⁸. Supine and standing lateral radiographs that were made one month after injury were used to determine correlations between the change in vertebral wedge height and pain. The median visual analog scale scores for back pain (possible range, 0 to 100 points, with 0 indicating no pain) with the patient in a supine position, in a standing position, and when standing erect were 13, 33, and 41, respectively, whereas the median change in wedge height from the supine to standing radiographs was 8% (p < 0.001), suggesting that changes in vertebral wedge height between these positions may yield important clinical information regarding the pathogenesis of pain resulting from fragility fractures of the spine.

The first systematic review to compare the efficacy and safety of balloon kyphoplasty and vertebroplasty for the treatment of vertebral compression fractures was published last year¹⁹. That review included three nonrandomized studies in which balloon kyphoplasty was compared with conventional medical therapy, thirteen case series of balloon kyphoplasty, one nonrandomized study in which vertebroplasty was compared with conventional medical care, and fifty-seven cases series of vertebroplasty. There were no randomized trials, and there was only one comparison study of the two techniques. The findings suggested that vertebroplasty and balloon kyphoplasty provide similar gains in pain relief, whereas for balloon kyphoplasty there is better documentation of gains in patient functionality. Regarding procedure safety, the level of cement leakage and the number of reported adverse events were significantly lower for balloon kyphoplasty than for vertebroplasty, although there was a good ratio of benefit to harm for both procedures. This Level-III evidence should be interpreted with caution and needs to be supplemented with randomized controlled trials.

With respect to burst fractures, we reviewed two randomized controlled trials, one of which compared two surgical procedures (fusion and nonfusion) and the other of which compared fusion with conservative management; these studies yielded insight but did not answer the question of best treatment. The first prospective trial was conducted to compare the results of posterior short-segment stabilization with and without fusion for surgically treated burst fractures of the thoracolumbar and lumbar spine in fifty-eight patients²⁰. The inclusion criteria were a kyphotic angle of 20°, a decrease of vertebral body height of 50%, or a canal compromise of 50%. The operative treatment included posterior reduction and instrumentation, with the fusion group (n = 30) receiving autogenous bone graft. After a mean duration of follow-up of forty-one months, neither the average loss of kyphotic angle nor the back pain scores were significantly different between the treatment groups. The radiographic parameters, however, were significantly different, with a lost correction of vertebral body height of 3.6% in the nonfusion group as compared with 8.3% in the fusion group (p < 0.002). The nonfusion group maintained segmental motion of 4.8°, compared with 1.0° for the fusion group (p < 0.001). Additionally, there was significantly less blood loss (303 compared with 572 mL) and operative time (162 compared with 224 minutes) in the nonfusion group. It would therefore appear that, over the short run, the advantages of instrumentation without fusion are the elimination of donor-site complications, the saving of motion segments, and the reduction in blood loss and operative time.

In a randomized trial from The Netherlands, spinal compression fractures without a neurological deficit were evaluated after a mean duration of follow-up of 4.3 years²¹. Eighteen patients received bisegmental posterior stabilization, and sixteen were managed with a conservative protocol. The patients in both groups were placed in a hyperextension brace for three months, and hardware was removed from the operatively managed patients between nine and twelve months postoperatively. The rate of follow-up was 94%. Both local and regional kyphotic deformity were significantly less in the operatively treated group (p < 0.0001). All functional outcome scores, including visual analog spine and pain scores, showed significantly better results in the operative treatment group (p = 0.02 to 0.03). There were no significant differences between the groups in terms of the rate of complications. The percentage of patients returning to their original jobs was significantly higher in the operative treatment group (85% compared with 38%), suggesting a benefit for surgery in this group of patients.

Pelvis
Morbidity after severe pelvic injury is often long-lasting, as shown in the two articles described in this section. A longitudinal single-cohort study of thirty-two operatively managed patients with unstable sacral fractures was performed to describe associated injuries and residual impairment in this severely injured population, providing insight into complications that are rarely well documented after pelvic surgery²². The mean Injury Severity Score was 27 (range, 9 to 57), and additional injuries occurred in 84% of the patients, with twenty-three of these injuries considered to be severe. After a minimum duration of follow-up of one year, 91% of the patients had sensory impairment, 63% had gait impairment, 38% had sexual impairment, and 21% had bowel or bladder dysfunction. Late impairments correlated with the severity of the injury and the presence of associated injuries, but not fracture characteristics. The surgical procedures were too varied to assess their affect on outcome.

Sacral decompression and lumbopelvic fixation in patients with spinopelvic dissociation is gaining a foothold in the treatment of unstable pelvic fractures. In a retrospective review of nineteen patients with highly displaced, comminuted, irreducible sacral fractures associated with neurologic deficits, eighteen patients met the criteria for a minimum one-year follow-up²³. All patients were managed with open reduction, sacral decompression, and lumbopelvic fixation. Sacral fractures healed in all eighteen patients without loss of reduction, and radiographic evaluation demonstrated the average sacral kyphosis improved from 43° to 21°. Fifteen patients (83%) had full or partial recovery of bowel and bladder deficits, although only ten patients (56%) had improved Gibbons neurologic scores. The average Gibbons score improved from 4 to 2.8 at a mean duration of follow-up of thirty-one months. Complete recovery of cauda equina function was more likely in patients with continuity of all sacral roots (86% compared with 0%; p = 0.00037) and incomplete deficits (100% compared with 20%; p = 0.024). As the wound infection rate of 16% reflected the most common complication, the benefit-to-harm ratio would appear to support lumbopelvic fixation for this severe condition.

Acetabulum
This section on the acetabulum is robust with outcome studies in which the authors attempted to present data on prognostic factors for these complex injuries, but the section begins with a study that aids diagnosis and ends with an efficacy study on heterotopic bone prevention.

In the study by Patel et al., the computed tomography scans and the anteroposterior and Judet radiographs of thirty patients were reviewed by six orthopaedic surgeons to help to determine a method of diagnosis that has good interobserver and intraobserver agreement in terms of the Letournel classification and important prognostic modifiers²⁴. There were eight yes-or-no questions regarding features on the pelvic radiographs and simple yes-or-no questions regarding the presence of prognostic modifiers (articular displacement, marginal impaction, incongruity, intra-articular fragments, and osteochondral injuries to the femoral head). This simple binary approach to the classification yielded an interobserver and intraobserver agreement that ranged from moderate to near-perfect (kappa = 0.49 to 0.88 and kappa = 0.57 to 0.88, respectively), which is a departure from similar assessments of other published fracture-classification systems. On the other hand, the prognostic modifiers yielded only slight to moderate interobserver and intraobserver agreement.

The first two articles covering the prognostic factors for acetabular fractures, both published in the British volume of The Journal of Bone and Joint Surgery, aimed to identify key features of posterior wall acetabular fractures and their treatment that correlate with outcome. The studies demonstrated both differences and striking similarities.

Bhandari et al. used a prospective database to identify 109 patients with an acetabular fracture associated with a posterior hip dislocation, all of whom were managed operatively within three weeks after the injury and were followed for at least two years²⁵. There were 38% simple patterns (32% of which were posterior wall patterns) and 61.5% associated patterns (41% of which were transverse plus posterior wall patterns). The mean age was forty-two years, and 72% of the patients were male. With use of multivariate analysis, the quality of reduction of the fracture was identified as the only significant predictor of radiographic grade, clinical function (as assessed with the Merle d'Aubigné score), and the development of posttraumatic arthritis (p < 0.001). All patients who lacked anatomical reduction (i.e., those who had a >1-mm step or gap detected on any one of three radiographic views of the pelvis) had development of radiographic signs of arthritis, whereas only 25.5% (twenty-four) of those with an anatomical reduction did so (p = 0.05). With use of univariable regression analysis, an interval of more than sixteen hours before relocation (p = 0.05) and the presence of femoral head damage (p = 0.03) were also associated with arthritis. No patient with an imperfect reduction had a good or excellent result at the time of the latest follow-up, whereas 70% of the patients with an anatomical reduction had a good or excellent result beyond twelve years postoperatively. Only nine patients (8%) required total hip arthroplasty.

In the second retrospective review, from Toronto, 128 patients who had been managed operatively for a fracture involving the posterior wall of the acetabulum were studied for the same reason²⁶. The mean age in this group was 41.6 years, and the fracture types included forty-four simple and eighty-four associated patterns. The Musculoskeletal Functional Assessment and Short Form-36 scores, the presence of radiographic signs of arthritis, and complications were assessed as a function of injury. The authors concluded that the patients had profound functional deficits compared with the normal population and that anatomical reduction alone was not sufficient to restore function. A posterior wall with more than three fragments (p = 0.005), marginal impaction (p = 0.01), and residual displacement of >2 mm (p = 0.001) were associated with the development of arthritis, which was related to poor function and the need for hip replacement. Hip replacement was required in 12.5% of the patients within 2.9 years after the injury.

Possible explanations for the discrepancy in findings between the studies include issues related to surgical technique, the determination of anatomicity, the use of different outcome instruments (with the latter study involving the use of validated tools), the number of patients with inadequate follow-up (forty-five in the first study, compared with forty-four in the second study), and the 17% prevalence of complete sciatic nerve palsy as well as the high number of associated bodily injuries (mean Injury Severity Score, 23.4) in the Toronto study. Collectively, however, they give us insight into some of the subtle and salient factors that determine prognosis.

In another retrospective report, Madhu et al. sought to determine the effect on outcome and arthritis that a delay to surgery might cause²⁷. In a study of 237 patients with displaced fractures of the acetabulum, surgical reduction was assessed, fractures were classified into elementary and associated patterns according to the Letournel system, and the time to surgery was documented. When the time to surgery was analyzed as a categorical variable (zero to five, six to ten, eleven to fifteen, sixteen to twenty, and more than twenty days), an anatomical reduction was found to be more likely when surgery was performed within fifteen days for elementary fracture patterns (p = 0.04) and five days for associated fracture patterns (p = 0.008). Significance in terms of functional outcome was first reached at fifteen days after surgery for elementary fracture patterns (p = 0.02) and at ten days for associated patterns (p < 0.0001). The authors pointed out that the organization of regional trauma services must be capable of satisfying these time-dependent requirements.

Borrelli et al. attempted to determine the correlation between muscular strength and patient function for fifteen patients in whom a displaced acetabular fracture was treated through an anterior ilioinguinal approach and then compared these results with those for a group of patients who were managed with a Kocher-Langenbeck posterior approach²⁸. After a mean duration of follow-up of forty-four months, the ilioinguinal group had an overall muscle strength deficit of 9% and an average Musculoskeletal Functional Assessment score of 17. After a mean duration of follow-up of twenty-four months, the Kocher-Langenbeck group had an overall strength deficit of 8% and an average Musculoskeletal Functional Assessment score of 22. Although the numbers in that study were very small, the authors found that the Musculoskeletal Functional Assessment scores did not differ significantly according to the surgical approach, the fracture pattern, gender, age, articular reduction, the radiographic grade, or the presence of heterotopic ossification. However, there was a correlation between hip extension/flexion work and maximum torque as well as between hip adduction work and maximum torque and the Musculoskeletal Functional Assessment score, leading the authors to conclude that an emphasis should be placed on hip muscle strength after the operative treatment of a displaced acetabular fracture in order to maximize outcome.

In the prospective, randomized trial by Karunakar et al., the effect of indomethacin was compared with that of a placebo for reducing the prevalence of heterotopic ossification²⁹.A total of 121 patients with a displaced fracture of the acetabulum that was treated through a posterior Kocher-Langenbeck approach were randomized to receive either indomethacin (75 mg sustained release) or a placebo once daily for six weeks. At three months postoperatively, Brooker grade-III or IV ossification had occurred in nine (15.2%) of fifty-nine patients in the indomethacin group and twelve (19.4%) of sixty-two patents in the placebo group; this difference was not significant. These results must be interpreted with caution, however, as the treatment groups differed significantly in terms of one important variable, hip dislocation, with 21% more hip dislocations in the placebo group. Furthermore, patient compliance with indomethacin was poor, with 43% of those who agreed to be tested having no detectable serum levels of indomethacin.

	Appendicular Skeleton

Top Introduction General Topics Clinical Axial Skeleton Appendicular Skeleton Appendix References

 
Upper Extremity
Shoulder and Proximal Part of the Humerus
While surgical treatment of most clavicular fractures was considered heresy only a decade ago, four articles that were published during the past year featured the surgical treatment of clavicular fractures. The most compelling work in this area has come from Toronto, with a compelling, prospective, randomized trial in which operative treatment was compared with nonoperative treatment for clavicular shaft fractures³⁰. In a well-controlled trial, 132 patients with a displaced midshaft clavicular fracture (with no cortical contact) who were seen at eight different centers were randomized to either sling treatment or a superior plating procedure. Constant scores and Disabilities of the Arm, Shoulder and Hand (DASH) scores were significantly improved in the operative fixation group at all time-points between six and forty-eight weeks (p < 0.01). The mean time to radiographic union was 28.4 weeks in the nonoperative treatment group and 16.4 weeks in the operative treatment group (p = 0.001). The nonoperative treatment group had more nonunions (seven compared with two, p = 0.042) and symptomatic malunions (nine compared with zero, p = 0.001). Complications in the operative treatment group included local irritation or prominence of the hardware (five patients), wound infection (three), and failure (one). One year postoperatively, the patients in the operative treatment group were more likely to be satisfied with the appearance of the shoulder (p = 0.001) and with the shoulder in general (p = 0.002) than were those who had been managed with a sling. This study presented solid evidence in support of operative management, but, because it did not stratify injury characteristics, it is impossible to know if there are subgroups that either do or do not benefit from surgery.

Despite the fact that the clavicle is such a superficial bone, open clavicular fractures are rare injuries, representing only 1.4% (twenty-four) of 1740 clavicular fractures that were collected in a prospective database at Harborview Medical Center (Seattle, Washington) over a thirteen-year period³¹. Twenty nonballistic fractures were evaluated with respect to injury characteristics and associations. The rate of head injury was 65%, and the rate of significant facial trauma was 55%. The rate of pneumothorax was 50%, and the rate of bilateral pneumothorax was 35%. The rate of cervical or thoracic spine injuries was 35%. The rate of scapular fractures was 40%, and the rate of additional ipsilateral upper extremity fractures was 30%. Interestingly, neurovascular complications were rare. All patients underwent débridement and irrigation, and fourteen of the twenty patients underwent open reduction and internal fixation. Of the fifteen patients with adequate follow-up, all had union.

In a study that seemed to corroborate the surgical outcomes found in the Canadian Trauma Group Clavicle Study described above, Collinge et al. reported on a series of eighty consecutive patients with a middle-third clavicular fracture or a painful nonunion of the clavicle that was treated with open reduction and internal fixation³². In contradistinction to the Canadian trial, which involved the use of a superior clavicular plating technique, this study highlighted the advantages of anteroinferior plating. Among the fifty-eight patients who had sufficient records and at least two years of follow-up, there was only a single nonunion. The mean healing time was 9.5 weeks for the patients who had had fracture fixations and 10.5 weeks for those who had had nonunion reconstructions. Except for the patients with neurologic injury trauma, the functional results were excellent and were comparable with normative data from age-matched controls. Complications included three infections, all of which resolved while the plate was left intact. In addition, two patients desired and underwent hardware removal. Five patients complained of minor irritation, and none complained of tenderness under bras or backpacks. The authors purported multiple advantages of the anteroinferior plate location, including the fact that drills are directed away from the important infraclavicular structures and that there is less likelihood of implant prominence.

A more challenging fracture with a track record for treatment failure is the unstable lateral clavicular fracture (Neer type II). The largest published series to date on the use of the clavicular hook plate (Stratec Medical, Oberdorf, Switzerland) for the treatment of this injury was a retrospective study of sixty-three patients from Finland³³. The Oxford score, the subjective part of the Constant score, and the subjective shoulder value were assessed for fifty-eight patients after a mean duration of follow-up of 3.6 years. The mean Oxford score was 15, indicating good function. The mean score for the subjective part of the Constant score was 32, compared with 34 on the control side (p = 0.003). The mean subjective shoulder value was 86%. Fifty-nine fractures united uneventfully. There was one delayed union, and there were three nonunions (two of which were thought to be the result of premature plate removal at two and three months); however, only one of these four complications required surgery. Additionally, there was one infection, one frozen shoulder, and three cases of late fracture medial to the plate after falls. The authors concluded that the hook plate provided a safe and effective solution for an otherwise difficult problem, but they emphasized the need for plate removal in all cases after healing.

In a prospective cohort study of 252 patients ranging from fifteen to thirty-five years of age who had an anterior glenohumeral dislocation that was treated with sling immobilization followed by a physical therapy regimen, instability developed in 55.7% of the shoulders within the first two years and in 66.8% by the fifth year³⁴. Patients with atraumatic instability, hyperlaxity with an atraumatic mechanism, and no Bankart lesion on magnetic resonance arthrography were excluded. A measurable degree of functional impairment was present at two years after the initial dislocation in most patients, and a significant risk for redislocation was noted among younger patients as well as among male patients (p < 0.01). Of the 134 patients who had a redislocation, 110 agreed to an operation, and of the twenty-four who declined, sixteen had another dislocation. The authors concluded that recurrent instability and functional deficits are common after primary nonoperative treatment of a traumatic anterior shoulder dislocation and that this injury warrants operative treatment in high-risk young male patients.

To evaluate the radiographic and clinical outcome (including the prevalence of recurrence) for displaced greater tuberosity fractures associated with traumatic anterior shoulder dislocation, Dimakopoulos et al. retrospectively studied thirty-four patients after a mean duration of follow-up of 4.8 years to assess the fixation of the tuberosity and the repair of any rotator cuff tear³⁵. The union rate was 97%. The result was excellent for twenty-five patients, very good for six, good for two, and poor for one. The average Constant score was 88.4, despite a 19.4% rate of neurologic injuries. All but one of the neurological injuries resolved after a mean of 3.5 months. No case of recurrence of dislocation was noted in any patient. Partial absorption of the tuberosity (with no clinical relevance) was detected in four cases. The authors reported that this treatment allows for early joint motion and yields excellent results in about three quarters of the patients.

Humeral Shaft and Elbow
Although orthopaedic surgeons have found more frequent indications for the operative treatment of humeral shaft fractures in the past decade, the results of three studies that were published during the past year generally supported nonoperative treatment, but with a couple of caveats. Two retrospective studies, one from Boston and one from Sweden, yielded similar conclusions^36,37. Both studies employed functional brace treatment until fracture-healing, and both identified a nonunion rate of 10%.

In the Swedish study, which included seventy-eight patients, the authors identified a trend toward more frequent nonunions in association with simple (Orthopaedic Trauma Association type-A) fractures³⁶. The nonunions that were operatively treated had an 88% union rate. Fifty patients completed long-term outcome examinations, and half reported full recovery after nonoperative treatment. However, none of the patients with a healed nonunion had full recovery after revision surgery. The authors warned that the acute simple fracture pattern in the proximal and middle thirds of the humerus may represent an indication for surgery.

In the Boston study, the main risk factor was the proximal one-third fracture, with four (29%) of fourteen such fractures having failed to heal³⁷. Interestingly, all of the proximal one-third fractures that did not heal were long spiral oblique fractures. There was no more than a 15° loss of shoulder or elbow motion among the patients who had healing of the fracture, and it is notable that angular deformity of >20° of varus occurred in three patients but was not thought to cause a functional problem.

Extending the theme of the previous two studies, Jawa et al. reported on fifty-one consecutive patients with a closed, extra-articular fracture of the distal one-third of the humeral diaphysis who were identified from an orthopaedic trauma database at two centers³⁸. Forty patients were followed for at least six months or until healing of the fracture; of these, nineteen had been managed with plate fixation and twenty-one had been managed with functional bracing. Among the operatively managed patients, one had loss of fixation and one had development of an infection. Three iatrogenic postoperative radial nerve palsies developed, one of which had not resolved at the time of follow-up three months after surgery. All operatively treated fractures healed with <10° of angular deformity, and only one patient lost 20° of shoulder or elbow motion. Among the nonoperatively managed patients, two had conversion to plate fixation because of concern with alignment, leaving only one patient with >30° of malalignment in any plane. Two patients lost >20° of elbow or shoulder motion, and two patients had development of skin breakdown during brace treatment. The authors concluded that operative treatment achieves more predictable alignment and a potentially quicker return of function but risks iatrogenic nerve injury and infection, whereas functional bracing can be associated with skin problems and varying degrees of angular deformity but usually is associated with excellent function and range of motion.

With the increased availability and improved technique of computed tomography, Doornberg et al. hypothesized that three-dimensional computed tomography reconstructions would improve diagnostic acumen with regard to our understanding and classification of distal humeral fractures³⁹. The computed tomography technique included imaging slices with <1.25-mm intervals, with the radius and ulna being subtracted from the image. Five observers evaluated thirty intraarticular fractures with respect to specific fracture characteristics and then classified the fractures according to two common schemes. Three-dimensional computed tomography improved both the intraobserver and interobserver reliability of classification and also improved the level of intraobserver agreement for all fracture characteristics from moderate to substantial, leading the authors to recommend its routine use for operative decision-making.

The second study on distal humeral articular fractures involved only partial articular fractures. In that study, seventy-nine patients with a mean age of forty-seven years were assessed retrospectively⁴⁰. This injury was found to occur most commonly in osteoporotic individuals, although there was a striking bimodal distribution in terms of age (more than eighty years and less than twenty years). The incidence among female patients over the age of sixty years was twice that among female patients under the age of sixty years. The mean age was 55.1 years for female patients, compared with 27.5 years for male patients. When the fracture occurred in young male patients, it was a more complex pattern from a higher energy mechanism. An associated radial head fracture occurred in 24% of the cases. Corroborating the findings of the previous study, preoperative classification from plain radiographs often underestimated the extent of the injury.

An anteromedial coronoid facet fracture, which results from a varus posteromedial rotational injury force, was treated in eighteen patients over a six-year period⁴¹. Twelve patients received treatment for the acute fracture, and six required revision after initial treatment elsewhere. All but three patients (two with a concomitant fracture of the olecranon and one with a second fracture at the base of the coronoid) had an avulsion of the origin of the lateral collateral ligament complex from the lateral epicondyle. Fracture patterns were characterized and instability was documented, with varus subluxation being observed in thirteen patients and posterior dislocation being noted in five patients. The initial treatment was operative in fifteen patients and nonoperative in three. The coronoid fracture was secured with various methods in eleven patients but was not repaired in the remaining seven patients. At the time of the latest follow-up, six patients had malalignment of the anteromedial facet of the coronoid with varus subluxation of the elbow, which was due to the fact that the fracture had not been fixed (four patients) or to the loss of fracture fixation (two patients). All six patients had a fair or poor result. The remaining twelve patients had good or excellent elbow function. The authors surmised that anteromedial fractures of the coronoid are associated with subluxation or complete dislocation of the elbow in most patients, making it imperative to stably fix the coronoid fracture.

Radial head replacement has become popular in recent years when the proximal part of the radius is "not reconstructable." Interpretation of this phrase in many circles seems to have included any and all comminuted radial head fractures, yet some have argued that successful execution of an anatomic and stable reduction of the native proximal part of the radius renders optimal results. In a prospective study from Cologne, Germany, the radiographic and clinical outcomes for twenty-three patients with a complex radial head fracture were evaluated at a median of two years after treatment of the injury with use of a new fixation device (Fragment Fixation System [FFS]; Orthofix, Bussolengo, Italy)⁴². The fixation device includes threaded wires that are self-drilling and self-tapping as well as a proximal shoulder that applies compression. Fourteen Mason type-III and eleven Mason type-IV fractures underwent open reduction and internal fixation. Seven patients had joint instability after repair, for which an articulating elbow fixator was used. All fractures united and, at the time of the latest follow-up, the functional elbow score was excellent for eight and good for four patients with a Mason type-III fracture and was excellent for five, good for three, and fair for three patients with a Mason type-IV fracture. Two patients had been lost to follow-up. Good elbow motion was achieved, leading to the conclusion that successful fixation of severe radial head fractures is possible and can lead to satisfactory functional results while preserving the native anatomy.

A retrospective review of sixty-three Monteggia fractures in adults was conducted to determine the prognosis of Bado and Jupiter classification variants on the basis of the mean 8.4-year follow-up of forty-seven patients who were managed operatively⁴³. Overall, the mean Broberg and Morrey score was 87.2 and the mean DASH score was 17.4. There was a significant correlation between the two scores (p = 0.01); however, twelve patients (26%) needed a second operation within twelve months. The salient finding of this study was that Bado type-II Monteggia fractures, and within this group, Jupiter type-IIa fractures, are frequently associated with fractures of the radial head and the coronoid process. These fractures were poor prognostic indicators, and these patients should be counseled accordingly regarding the possible need for further surgery.

Distal Part of the Radius
The Massachusetts General Hospital upper extremity group conducted a study of distal radial fractures to investigate the added value of the three-dimensional computed tomography scan in diagnosis⁴⁴. Four independent observers evaluated computed tomography and plain radiographic images of thirty intra-articular fractures of the distal part of the radius for the presence of a fracture line in the coronal plane, impacted central articular fragments, the presence of comminution, and the number of fracture fragments. These observations were then compared against intraoperative findings. Three-dimensional computed tomography significantly improved intraobserver agreement, but not interobserver agreement, regarding the presence of coronal plane fracture lines and central articular fragment depression. It improved both intraobserver agreement regarding the presence of articular comminution and interobserver agreement for determining the number of articular fracture fragments. Perhaps most importantly, the addition of three-dimensional computed tomography influenced treatment recommendations, resulting in a significantly greater number of decisions for an open approach and a combined dorsal and volar exposure. The authors concluded that three-dimensional computed tomography improves both the reliability and the accuracy of radiographic characterization of articular fractures of the distal part of the radius and influences treatment decisions.

Mackenney et al. tried to identify the predictors of fracture instability and to construct a method of predicting the radiographic outcome (instability) by analyzing data on 4000 distal radial fractures that had been collected over a five-year period⁴⁵. Predictors of early and late instability and malunion differed according to the displacement of the fracture at the time of presentation. Patient age, metaphyseal comminution, and ulnar variance were the most consistent predictors of radiographic outcome, and dorsal angular deformity was not found to be a significant predictor. Interestingly, the classification of a patient as independent (that is, able to go shopping alone) was predictive of malunion in displaced fractures. A formula that stratifies the risk of each variable was presented to allow for the prediction of loss of reduction and malunion in an effort to aid the surgeon's decision-making regarding the primary treatment of distal radial fractures.

The purpose of a Swedish randomized trial was to compare wrist-bridging and non-wrist-bridging external fixation for displaced distal radial fractures in the elderly⁴⁶. Thirty-eight patients (mean age, seventy-one years) were randomized to treatment with a bridging external fixator or a nonbridging external fixator. At the time of follow-up, there was no significant difference in DASH or visual analog pain scores. There was also no significant difference in terms of range of motion, grip strength, or patient satisfaction. Although volar tilt and radial inclination were the same in both groups, the nonbridging group maintained a significantly better radial length at one year. The authors concluded that there was no clinically relevant benefit to nonbridging external fixation in this group of patients, but there should be caution with interpretation given the small numbers.

Twenty patients with a fracture of the dorsal articular margin of the distal part of the radius with dorsal radiocarpal subluxation (a reverse Barton fracture) were evaluated at a mean of thirty months after open reduction and internal fixation⁴⁷. Surgical findings included major volar injuries, including two torn volar ligaments, ten rotated volar marginal lip fractures, and six volar articular impactions. Fourteen of the twenty patients also had impacted central articular fragments. Eighteen patients underwent surgical reconstruction of the articular surface and application of a dorsal buttress plate. Nineteen fractures healed without substantial loss of alignment. The final average wrist and forearm motion was 59° of flexion, 56° of extension, 87° of pronation, and 85° of supination. The mean grip strength was 85% of that of the contralateral, uninjured hand. The final functional result according to the system of Gartland and Werley was rated as excellent or good for eighteen patients and fair for two. The average modified Mayo wrist score was 75 and the average DASH score was 15 points, leading to the conclusion that despite the complexity of these injuries, satisfactory wrist function can be achieved with operative treatment in most patients.

In a surgical technique article, twenty-three skeletally mature patients were evaluated at a mean thirty-eight months after an intra-articular osteotomy for the treatment of a malunion of the distal part of the radius⁴⁸. The indications for the osteotomy included dorsal or volar subluxation of the radiocarpal joint in fourteen patients and articular incongruity of 2 mm as measured on a posteroanterior radiograph in seventeen patients. Six patients had a combined intra-articular and extra-articular malunion, and the mean step-off or gap of the articular surface prior to the operation was 4 mm. The average interval from the injury to the osteotomy was six months. With regard to reoperations, one patient had partial wrist arthrodesis because of radiocarpal arthrosis, three patients had additional surgery because of dysfunction of the distal radioulnar joint, and one patient had tendon transfer because of a rupture of the extensor pollicis longus. After healing of the osteotomy site, the mean articular incongruity averaged 0.4 mm. Clinical results included a final grip strength that averaged 85% of that on the contralateral side. The rate of excellent or good results was 83% according to the rating system of Fernandez and that of Gartland and Werley. The authors pointed to the comparable results with extra-articular osteotomies of the distal part of the radius and noted that an intra-articular osteotomy may limit the need for a salvage procedure such as arthrodesis.

Lower Extremity
Proximal Part of the Femur
Perhaps one of the most significant movements in fracture care over the past decade has been the trend away from internal fixation of femoral neck fractures in the elderly. This trend seems to have resulted not from the relative performances of a healed fracture with a well-perfused femoral head and a hemiarthroplasty or total hip arthroplasty but from the occurrence of complications such as nonunion and osteonecrosis in >30% of patients. A Cochrane review of all randomized controlled trials that have compared internal fixation with arthroplasty for the treatment of intracapsular femoral fractures in adults did not lend clear-cut evidence for this modern trend away from internal fixation⁴⁹. After screening for trial quality, seventeen randomized, controlled trials involving 2694 participants were included. The results of the review demonstrated significant decreases in the duration of surgery, the amount of blood loss, the need for blood transfusion, and the rate of deep wound infection for internal fixation relative to arthroplasty; however, arthroplasty had a lower reoperation rate in comparison with fixation. There were no clear differences in terms of the length of hospital stay, mortality, or return to residence status. There was a suggestion from some studies that arthroplasty was associated with lower rates of pain and higher function. The authors concluded that future trials need to focus better on measuring functional outcome to decipher appropriate patient selection for each procedure.

The complexity of this debate on the best method for the treatment of intracapsular hip fractures is underscored by diverging conclusions of a meta-analysis from Sweden in which fourteen randomized controlled trials involving 2289 patients between 1966 and 2004 showed that primary arthroplasty leads to significantly fewer major complications of surgery, including infection, and fewer reoperations⁵⁰. As in the previous review, however, there was no significant difference in mortality at thirty days or one year between the two groups. The authors stated that most of the studies demonstrated better function and less pain after primary arthroplasty.

The debate has shifted during the past couple of years to whether either a hemiarthroplasty or total hip arthroplasty provides a superior treatment option. In one prospective randomized study comparing these treatment options, eighty-one patients who had been mobile and had lived independently before sustaining a displaced fracture of the femoral neck were randomized to each of these two treatment groups⁵¹. The mean age of the patients was seventy-five years, and the mean duration of follow-up was three years. Total hip arthroplasty conferred superior short-term clinical results as demonstrated by greater walking distance (3.6 compared with 1.9 km; p = 0.039) and a lower Oxford hip score (p = 0.033). Furthermore, twenty of thirty-two living patients in the hemiarthroplasty group had radiographic evidence of acetabular erosion at the time of the final follow-up. Two patients required revision to a total hip arthroplasty, and three additional hips had acetabular erosion severe enough to warrant revision. In contrast, there were three dislocations in the total hip arthroplasty group and none in the hemiarthroplasty group. Overall, the authors concluded that a total hip arthroplasty conferred a better functional outcome with fewer short-term complications.

Another meta-analysis focused attention on the use of cement (as opposed to press-fit fixation) in the treatment of femoral neck fractures with arthroplasty⁵². Seventeen trials involving 1920 patients were included, although there were few direct comparisons in trials in which the components were a controlled variable. In comparison with uncemented prostheses, cemented prostheses were associated with less pain at one year or later and demonstrated a tendency toward better mobility. No significant difference in surgical complications was found. In two trials involving 232 patients, hemiarthroplasty without cement was compared with total hip replacement. Both studies demonstrated increased pain for the uncemented prosthesis, and one study demonstrated better mobility and a lower long-term revision rate for patients managed with total hip arthroplasty. In two trials involving 214 participants, hemiarthroplasty with cement was compared with total hip replacement. Both trials demonstrated little difference between prostheses except a slightly longer surgical time and a tendency toward better function in the total hip arthroplasty group. The authors of that meta-analysis concluded that cementing a prosthesis in place may reduce postoperative pain and lead to better mobility.

In summary, arthroplasty with cement seems to have emerged as the preferred solution for the treatment of femoral neck fractures in the elderly. It is important to realize, however, that these studies ultimately are comparing any type of internal fixation with any type of arthroplasty (over four decades) as well as the myriad of surgical techniques and approaches associated with such operations. These variables leave plenty of room for speculation as to whether a native hip with a well-fixed fracture without complications would out-perform an arthroplasty over the long term. Indeed, future studies need to focus on this remaining question to sort out appropriate roles for these various techniques.

There has been no debate on the use of arthroplasty as opposed to internal fixation for the treatment of extracapsular hip fractures; rather, the debate has centered around types of internal fixation. In a randomized controlled trial from Nepal, published in the British volume of The Journal of Bone and Joint Surgery, the sliding hip screw was compared with three months of external fixation for the treatment of intertrochanteric hip fractures in sixty-seven patients with a mean age of sixty-six years⁵³. With the two groups being matched according to age and gender, the results revealed that the time to surgery, the duration of surgery, the amount of blood loss, the length of hospital stay, and the cost of treatment were all significantly higher in the sliding hip screw group. There was no significant difference at six months with regard to the time to union, the range of hip motion, the mean Harris hip score, or the quality of reduction of the fracture. Although pin-track infections occurred in 60% of the patients with external fixators, there was no difference in the rate of deep infection. Interestingly, all patients managed with external fixation had local anesthesia during the operation, whereas all patients in the internal fixation group underwent regional anesthesia. These results may support the idea that external fixation could be the most viable option in societies in which resources are scarce. These results also corroborate previous evidence on the viability of external fixation for the treatment of intertrochanteric hip fractures and collectively seem to indicate that a quest for some internal fixation device that maintains femoral neck length and neck-shaft angle may indeed be the goal.

Although the sliding hip screw has remained the gold standard for the treatment of intertrochanteric fractures, there has been greater recognition of poor outcomes in the elderly population. A greater understanding of the variables that portend a poor prognosis is therefore critical. In the third study in as many years to point to the integrity of the lateral cortex of the proximal part of the femur as being the critical variable in surgical outcome, 214 consecutive patients with an intertrochanteric fracture that was treated in Denmark with a 135° sliding compression hip screw with a four-hole side-plate were studied⁵⁴. Only five (3%) of 168 patients with an intact lateral femoral wall postoperatively underwent a reoperation within six months, whereas ten (22%) of forty-six patients with a fractured lateral femoral wall underwent reoperation (an eight times higher risk). Furthermore, 74% of the detected fractures of the lateral femoral wall occurred during the operative procedure itself, with such a fracture occurring in 31% of the ninety-nine patients with a fracture of the lesser or greater trochanter. It is notable that the implant position (tip apex distance) was also found to have an effect (although lesser) on the reoperation rate but a fractured lesser trochanter did not.

A relatively high rate of reoperation has been associated with Gamma nailing of trochanteric fractures. In a study involving a consecutive series of 554 patients, the outcomes for fifty-two patients who had a reoperation because of the failure of a Gamma nail were compared with those for the remaining 502 patients who had no reoperation⁵⁵. The most common reasons for a reoperation were a new fracture around the implant (seventeen), local pain after a healed fracture (eleven), nonunion (nine), and cut-out (eight). A second reoperation was required in nine (17%) of the fifty-two patients. The mortality was significantly lower at thirty days and at one to five years in the patients who underwent a reoperation, and there were no significant differences in independent walking ability or pain, indicating that reoperation did not lead to a worse clinical outcome. As the rate of loss to follow-up was 30% and the patients who had a reoperation