INTRODUCTION
Displaced midshaft clavicle fractures are common injuries, accounting for about 2.4%–4% of all adult fractures [1]. While nonsurgical management is appropriate for minimally displaced fractures, operative intervention is generally favored in cases of marked displacement, comminution, or high functional demand [2,3]. Plate fixation has emerged as the standard of care in such cases, offering predictable union rates, anatomic alignment, and durable outcomes [4-6]. However, the optimal plate orientation—specifically, superior versus anteroinferior plating—remains a subject of ongoing debate [7-9].
Superior plating offers increased resistance to axial loading but may be associated with greater hardware prominence and soft-tissue irritation [10,11]. Anteroinferior plating, on the other hand, allows for improved soft-tissue coverage and a reduced risk of plate-related irritation [12-14]. Several comparative studies have evaluated these differences; however, their conclusions have been inconsistent, reflecting variability in sample size, study design, follow-up duration, and outcome measures [15-19].
A 2024 meta-analysis by Hornung et al. [20] found no significant differences in union rates, functional outcomes, or complication rates between these surgical techniques. However, their study was limited by several methodological shortcomings. First, the included studies often featured indirect comparisons, with plate orientation assessed as a secondary or subgroup variable across heterogeneous cohorts. Second, pooled analyses exhibited considerable statistical heterogeneity and relied on variable-quality evidence, as reflected by low certainty on the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) scale. Third, the search strategy excluded trials published after March 2022, and the actual analysis incorporated studies published only through 2017, therefore omitting a substantial number of newer studies [20].
In response, we conducted a systematic review and meta-analysis limited to direct comparative studies of ranging levels of evidence (level I randomized controlled trials [RCTs] to level II/III observation cohort studies) evaluating superior versus anteroinferior plating in adult patients with displaced midshaft clavicle fractures. Primary outcomes included time to fracture union and nonunion rates, along with functional outcomes and overall complication rates; secondary outcomes included hardware removal and infection rates. By restricting the analysis to head-to-head comparisons and incorporating the latest data, this study sought to determine whether plate orientation meaningfully affects patient outcomes.
METHODS
This study did not involve human participants or identifiable patient data; therefore, institutional review board approval and informed consent were not required.
Search Strategy and Criteria
This study was registered in PROSPERO (CRD420251139555). A comprehensive literature search was conducted from database inception through August 2025 across five databases: PubMed, the Cochrane Library, Scopus, Embase, and Google Scholar. A Boolean keyword search combining MeSH and free-text terms was performed using the following syntax: (“clavicle” OR “clavicular”) AND (“midshaft” OR “mid‑shaft” OR “diaphyse*”) AND (“plate*” OR “plating”) AND (“superior” OR “anteroinferior” OR “anterior”).
Inclusion and Exclusion Criteria
Only direct comparative studies evaluating anteroinferior versus superior plating for mid-clavicular fractures in adults were included. Eligible studies included RCTs and nonrandomized prospective or retrospective comparative studies. At least one of the following outcome measures were required: Constant-Murley score; Disabilities of the Arm, Shoulder, and Hand (DASH)/QuickDASH; time to union; nonunion; infection (superficial or deep); plate removal; refracture; or neurovascular injury. Articles were required to be in English or to have an English translation available. Studies such as case reports and those with indirect or noncomparative designs were excluded. Additional exclusion criteria included inaccessible full texts or lack of an English-language version. Two reviewers (JL and FAT) independently screened all titles and abstracts using Rayyan.ai, with discrepancies resolved by a third reviewer (BH) and group consensus. Three independent researchers (JL, FAT, TL) performed data extraction using a standardized collection sheet.
Assessment of Study Quality
Certainty of evidence was assessed using the GRADE criteria by two independent authors (JL and TL) [21]. Risk of bias was evaluated using the Cochrane Risk of Bias 2.0 (RoB 2.0) tool for RCTs and the Risk of Bias in Nonrandomized Studies of Interventions (ROBINS-I) tool for nonrandomized studies [22,23]. Discrepancies were adjudicated by a senior reviewer (BH).
Statistical Analysis
All analyses were conducted using Review Manager version 5.4 (Cochrane). For dichotomous outcomes, pooled odds ratios (ORs) with 95% CIs were calculated using a random-effects model to account for expected heterogeneity. Meanwhile, continuous outcomes were analyzed as mean differences (MDs) under a random-effects model. Subgroup analyses were performed to compare results by plating orientation and to assess overall pooled effects. Heterogeneity was assessed using the chi-squared test and quantified with the I2 statistic. P<0.05 was considered statistically significant.
RESULTS
Article Selection/Study Characteristics
The initial search yielded 1,563 studies across the five chosen literature databases through August 24, 2025. After removing 809 duplicates, 754 studies remained for independent screening. Of these, 739 were excluded following title and abstract review. Fifteen full-text articles were then assessed for eligibility and three were excluded, two due to missing outcomes data and one due to an indirect comparison. Twelve studies ultimately met the study inclusion criteria and were incorporated into the final meta-analysis (Fig. 1). Although statistical heterogeneity was minimal across pooled analyses (I2=0%), clinically relevant variability was present among the included studies, including differences in patient age distributions, fixation techniques, implant designs, follow-up duration (ranging from 6 months to over 6 years), and thresholds for reoperation (e.g., elective hardware removal). The included studies are summarized in Table 1 [2,4,5,7,8,10-16].
Functional Outcomes
Four studies (n=219) reported Constant-Murley scores at ≥2 years. No significant difference was observed between superior and anteroinferior plating (MD=−1.19; 95% CI, −3.18 to 0.81; P=0.24; I2=0%). Across individual studies, mean Constant-Murley scores ranged from 82.67–97.27 points in both groups. Similarly, four studies (n=302) assessed DASH scores, with no significant difference between groups (MD=1.62; 95% CI, −0.46 to 3.70; P=0.13; I2=0%); individual study DASH mean scores ranged from 1.4–7.97 points (Fig. 2).
Radiographic, Surgical, and Union-related Outcomes
Four studies (n=220) evaluating time to union demonstrated no difference between superior and anteroinferior fixation (MD=0.41 weeks; 95% CI, −0.60 to 1.43; P=0.42; I2=0%). Time-to-union individual study means ranged from 11.28–23.6 weeks. The risk of nonunion was reported in six studies (n=378) without a significant difference (OR=2.42; 95% CI, 0.59–9.94; P=0.22; I2=0%). Included studies reported between zero and six nonunion events. Rates of refracture rates following plate removal were also similar (OR=1.18; 95% CI, 0.27 to 5.05; P=0.83; I2=0%), with reported refracture events ranging from zero to two (Fig. 3).
Complications
Nine studies (n=831) reported plate-removal rates, with no significant difference between superior and anteroinferior fixation (OR=1.16; 95% CI, 0.82–1.65; P=0.41; I2=0%). Seven studies (n=590) evaluated infection rates, including both superficial and deep surgical-site infections, and likewise showed no significant difference (OR=0.81; 95% CI, 0.32–2.06; P=0.66; I2=0%). Neurovascular injury was rarely reported and did not differ between groups (OR = 5.35; 95% CI, 0.25–116.31; P=0.29). Therefore, between-study heterogeneity could not be assessed for this outcome because only one study contributed events, with zero events reported in the remaining studies. Individual study-reported event ranges for each complication outcome were as follows: 0–34 plate removals, zero to two neurovascular injury events, and zero to three infections (superficial or deep) (Fig. 4).
Certainty of Evidence and Risk of Bias Assessment
Risk of bias was evaluated using RoB 2.0 for randomized trials and ROBINS-I for observational studies. Most RCTs demonstrated a low risk of bias, though two had concerns regarding randomization and selective reporting. This reduces confidence in the precision of our effect estimates, despite showing no impact in the pooled analysis. Among observational studies, most were rated moderate risk, with a few serious concerns related to participant selection and incomplete outcomes data (Fig. 5). This limits causal inference of our pooled analysis and raises important considerations for factors beyond plate orientation—such as fracture severity, implant characteristics, and postoperative care—that may have influenced our results.
The certainty of evidence was evaluated using the GRADE framework across five domains: risk of bias, inconsistency, indirectness, imprecision, and publication bias. Functional outcomes (Constant-Murley, DASH), time to union, nonunion, refracture, and infection were downgraded for imprecision and, in some cases, risk of bias due to modest sample sizes and rare events. In particular, nonunion and refracture outcomes were downgraded for very serious imprecision, as wide CIs encompass clinically meaningful benefit and harm, which substantially limits the strength our conclusions, despite pooled estimates suggesting no difference between techniques. Neurovascular injury outcomes lacked sufficient data for a meaningful GRADE assessment. Despite these downgrades, the overall body of evidence demonstrated consistent directionality and no clinically meaningful differences between plating techniques. Publication bias appeared minimal based on funnel plot symmetry (Fig. 6). Accordingly, the certainty of evidence was rated moderate to low for most outcomes, reflecting that, while equivalence between techniques is supported, confidence in rare-event outcomes remains limited, and conclusions should be interpreted cautiously (Table 2).
DISCUSSION
This meta-analysis found no significant differences between superior and anteroinferior plating across functional, radiographic, surgical, and complication-related outcomes. These results are consistent with prior individual studies evaluating the two surgical techniques. For instance, Sangiorgio et al. [10] conducted a prospective study with long-term follow-up and found no significant differences in union rates, functional outcomes (Constant-Murley and DASH), return to sport, or implant removal between superior and anteroinferior groups. Similarly, Nolte et al. [8] reported high union rates and comparable complication profiles between approaches in a cohort of over 100 patients. A prior meta-analysis by Nourian et al. [24] likewise found no significant differences in union, malunion, nonunion, or implant failure, although it suggested that superior plating may carry a higher rate of symptomatic hardware and implant removal. Unlike the most recent 2024 meta-analysis by Hornung et al. [20], which relied largely on indirect subgroup analyses and excluded trials published after 2017, the present study incorporates recent direct comparisons, standardized outcome measures, and increased follow-up. By focusing on direct comparisons and including newer, higher-quality evidence, this analysis provides a more contemporary synthesis that addresses key methodological gaps in the existing literature.
Our collective findings support the notion that, while superior plating may demonstrate greater mechanical strength in vitro, its biomechanical advantages do not translate into superior clinical outcomes. Cadaveric studies by Celestre et al. [25] and Toogood et al. [9] showed that superior plates withstand higher loads before failure; however, in vivo, these differences appear clinically negligible, particularly when modern pre-contoured plates and contemporary fixation techniques are employed to achieve stability. This emphasizes that clinical decision-making should rely on patient-specific and practical considerations, as opposed to theoretical strength differences.
An important consideration in clinical decision-making is patient comfort and hardware tolerance. Although anteroinferior plating has been hypothesized to reduce implant-related symptoms, this meta-analysis did not reveal a statistically significant reduction in plate removal or hardware irritation when using anteroinferior plating compared to superior plating. Sangiorgio et al. [10] similarly observed high overall implant removal rates in both groups, largely driven by patient discomfort, but found no significant difference by orientation. Other studies, such as those by Formaini et al. [4] and Virtanen et al. [26], have noted that hardware prominence is multifactorial, influenced by body habitus, implant design, surgical technique, and plate position. Recent reports, including those by Salazar et al. [19] and Ryan et al. [27], suggest that newer pre-contoured anterior plating may reduce irritation and removal rates, and registry data indicate that removal rates for superior plates are declining, likely reflecting evolution in implant design and surgical practice.
This study has several strengths, including its focus on direct comparisons, inclusion of the most recent literature, and comprehensive evaluation of clinically relevant outcomes. However, several limitations must be acknowledged. Most included studies were nonrandomized, and variations in fixation technique, implant type, and postoperative protocols may introduce potential confounding. Assorted definitions of hardware-related symptoms and thresholds for implant removal were used, limiting comparability of pooled data. Stratification of infection, whether superficial or deep, was not possible due to few studies specifying the difference. Additionally, sample sizes were modest, and follow-up durations ranged among the studies, possibly underestimating late complications or reoperations.
Overall, this study found no significant differences in clinical outcomes between superior and anteroinferior plating for midshaft clavicle fractures. While no significant differences were identified, equivalence between superior and anteroinferior plating should be interpreted alongside inherent variability in fixation technique amongst surgeons, differences in implant design, individual patient body habitus, and—among the included studies—heterogenous criteria for elective hardware removal. Nevertheless, these findings support that plate orientation can be guided by patient anatomy, surgeon preference, and intraoperative considerations rather than perceived differences in outcomes. Future studies should incorporate larger, prospective cohorts, considering standardized functional and radiographic outcome measures with longer follow-up durations.
CONCLUSIONS
Superior and anteroinferior plating of midshaft clavicle fractures yielded comparable union rates, functional outcomes, and complication profiles. Both plate orientations provide equivalent outcomes. The surgical approach should depend on fracture morphology, surgeon preference, and patient-specific factors.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
