Importance of the Topic

Tibial shaft fractures are the most common of long bone fractures [7, 13], occurring in 17 per 10,000 persons annually [6] and this rate is growing as low-and middle-income countries become increasingly motorized [11]. Tibial shaft fractures typically are high-energy injuries, and often are complicated by deep infection, nonunion, malunion, and compartment syndrome [6]. These complications can lead to prolonged pain and disability, and they consume substantial healthcare resources [1, 10].

Current surgical options include external fixation, plate fixation, and intramedullary nailing. Survey data indicate nearly 90% of orthopaedic trauma surgeons prefer intramedullary nailing for both open and closed tibial shaft fractures, but there is considerable variability regarding the preferred surgical approach, method of interlocking, nail material, and decision to ream the medullary canal [2, 5].

This Cochrane review determined whether different methods and types of intramedullary nailing were associated with unique benefits and harms when treating adults with tibial shaft fractures. Five different comparisons of interventions were assessed across 11 separate trials in 2093 participants (2123 fractures). The authors found insufficient evidence to determine the effects of treatment between most of the comparisons of interest, with the exception of reamed versus unreamed nailing. Reaming was associated with a lower risk of implant failure, and incidence of major reoperations related to nonunion in closed compared to open fractures (low-quality evidence).

Upon Closer Inspection

Unreamed nails have been postulated to preserve the intramedullary blood supply, but reaming is thought to deposit autograft bone at the fracture site while improving cortical contact at the bone-nail interface, increasing the stability of the construct [3, 8]. The majority of the trials included in the review compared these two interventions, but the results were dominated by one international, multicenter trial. The Study to Prospectively Evaluate Reamed Intramedullary Nails in Patients with Tibial Fractures (SPRINT) contributed 63% of the overall study review population (1319/2093), and 83% of the population from six trials addressing reamed versus unreamed nailing (1319/1588) [15].

It is important to account for imbalances in sizes of studies when conducting meta-analyses. To do so, the authors performed sensitivity analyses in which they switched from random-effects to fixed-effects statistical models [4]. Fixed-effect models assume constant treatment effects across each of the studies included, such that studies have effect sizes (and weight) directly proportional to their number of patients, an approach that tends to discount the results of smaller studies relative to large ones. By contrast, random-effect models aim to estimate mean distributions of effects, which minimizes the influence of individual large studies and leads to more conservative summary estimates [4, 12]. The results of this review did not differ between fixed- and random-effect models, indicating that the results are robust and protected from the influence of the larger studies.

Take-home Messages

This Cochrane review found no important differences in rates of reoperations and complications between reamed and unreamed intramedullary nailing. The only prespecified subgroup analysis with sufficient data to be assessed was open versus closed fracture, which suggested that reamed nailing is more likely to reduce the incidence of major reoperations related to nonunion in closed fractures, where as in open fractures there was no difference open. A recent report of prognostic factors from SPRINT published subsequent to the Cochrane review [14] found that open fractures had a higher risk than closed fractures of requiring a secondary intervention or developing a compartment syndrome when treated with reamed nailing, but not when treated with unreamed nailing.

According to the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) approach [9], there is moderate quality evidence to suggest no substantial difference between reamed and unreamed nailing for rates of reoperation, nonunion, pain, deep-infection, malunion, or compartment syndrome. With respect to the other comparisons (one versus two distal locking screws, interlocking versus expandable nail, paratendenous versus transtendinous approach), the studies included were of low quality and subject to bias, which limited the investigators’ ability to confidently pool data and draw meaningful conclusions. Additional large, high-quality studies are needed and to definitively examine the role of locking versus nonlocking implants, type of locking used, the number of locking screws required and the ideal approach for nail insertion. Particularly with the advent of newer nail insertion techniques (infra verses suprapatellar nailing), this evidence will be forthcoming as multicenter, randomized controlled trials are underway.