Surgical Management for Olecranon Fractures in Adults: A Systematic Review and Meta-analysis

Materials and Methods

The authors conducted a systematic review and meta-analysis in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses statement^32,33 and the Cochrane Collaboration's handbook for systematic reviews of interventions.³⁴

Literature Review

The authors conducted a detailed search to identify studies examining surgical management strategies for fractures of the olecranon. Their electronic search included 5 databases: MEDLINE, Embase, SPORTDiscus, CINAHL, and the Cochrane Library. Search strategies were developed collaboratively by all of the authors and were conducted without any restrictions on type of publication or time using key words and medical subject heading terms such as “olecranon,” “elbow fracture,” “tension-band wiring,” and “plate fixation” (Table A, available in the online version of the article). The search is current to November 20, 2018. The authors hand searched the references of included studies and reviews to identify articles not captured by electronic search criteria.

Table A: Search strategy for the Medline database

Results

The initial literature search yielded 3760 articles. After de-duplication and title and abstract screening, 100 articles remained. After full-text screening, 24 articles were included in the systematic review^{12–17,19–22,30,31,39–50} and 10 were included in the meta-analysis^{21,31,39,42,43,45–47,49,50} (Figure 1; Table B, available in the online version of the article). Cochrane risk of bias for RCTs indicated that studies were of relatively low to moderate quality, with recent RCTs^15,16,30,31 having moderate quality (Tables C–E, available in the online version of the article). All studies showed high risk of bias for patient blinding.

Figure 1: Search strategy.

Table B: Characteristics of included studies

Table C: Cochrane risk of bias for randomized controlled trials

Table D: Newcastle-Ottawa for prospective cohort studies

Table E: MINORS for retrospective cohort studies

A systematic review of PF vs TBW, involving 639 patients across 7 observational studies^{42,43,45–47,49,50} and 3 RCTs,^21,31,39 showed some clinical heterogeneity regarding reporting of outcomes. Patients had a mean age of 40.7 years at presentation and 32.6% were female. Functional outcomes were reported in 6 observational studies and 1 RCT (Table 1). Average functional outcome after both TBW and PF was either good or excellent in all studies according to Mayo Elbow Performance Scores, with all but 1 study reporting a mean elbow flexion arc greater than 130°.⁴² One cohort study (n=48) found significantly less terminal extension (−8.6º±7º vs −3.5º±9.3º) after hook-plate fixation⁴⁹; no other studies reported functional differences between TBW and PF.

Table 1: Functional Outcomes

Complication rate was reported in 7 cohort studies and 3 RCTs. There were 155 of 270 (57%) patients with complications who underwent TBW and 81 of 369 (22%) patients with complications who underwent PF (Table 2). The meta-analysis revealed significantly lower complication rates with PF compared with TBW (relative risk, 0.48; 95% confidence interval, 0.36–0.64; P<.01; I²=16%; Figure 2). These results were robust to subgroup analysis for complication rate comparing RCTs with observational studies (P=.45).

Table 2: List of Complications

Figure 2: Forest plot of tension-band wiring vs plate fixation for outcome of complications. Abbreviations: CI, confidence interval; MH, Mantel–Haenszel.

Hardware removal was reported in 7 cohort studies and 1 RCT. There were 41 of 332 (12%) patients with hardware removal who underwent PF and 79 of 236 (33%) patients with hardware removal who underwent TBW (Table 2). The meta-analysis revealed lower hardware removal with PF compared with TBW (relative risk, 0.36; 95% confidence interval, 0.25–0.50; P<.01; I²=0%; Figure 3). In addition, these results were robust to subgroup analysis for complication rate comparing RCTs with observational studies (P=.54).

Figure 3: Forest plot of tension-band wiring vs plate fixation for outcome of hardware removal. Abbreviations: CI, confidence interval; MH, Mantel–Haenszel.

Three RCTs and 1 quasi-RCT reported outcomes following other methods of fixation (Table 3).^14–16,30 An RCT by Liu et al¹⁵ compared a modified cable pin system tension band (n=30) with standard TBW (n=32) and found that the modified cable pin system tension band was associated with significantly shorter fracture healing time (9.73±2.02 vs 11.13±2.21 weeks), significantly fewer complications (1 vs 7), and significantly greater Mayo Elbow Performance Score (88.67±6.42 vs 80.78±11.99). Lu et al¹⁶ published an RCT comparing TBW through 2 cannulated screws (n=42) with standard TBW (n=46). They reported that cannulated screw use was associated with significantly shorter fracture healing time (11.4±1.2 vs 12.6±1.8 weeks), significantly less hardware removal (0 of 42 vs 10 of 46), and significantly greater Mayo Elbow Performance Score (87.9±6.0 vs 83.67±6.6).¹⁶ An RCT by Chen et al³⁰ comparing novel olecranon memory connector fixation (n=20) with locking PF (n=20) reported that olecranon memory connector fixation was associated with greater Mayo Elbow Performance Score (86.7±12.5 vs 79.8±12.3) and no difference in complication rate. Finally, the quasi-RCT by Ahmed et al¹⁴ examining cancellous screw TBW (n=15) vs standard K-wire TBW (n=15) found that screw TBW was associated with fewer re-operations for removal of prominent hardware (0 of 15 vs 8 of 15) and no functional differences.

Table 3: Other Elbow Fixation Techniques

Discussion

This systematic review and meta-analysis found that among patients who have olecranon fractures and are eligible for surgery, those who undergo PF show significantly fewer complications and hardware removals compared with those who undergo TBW. The current analysis focused on TBW and PF because they represent the most common methods of olecranon fixation; however, several novel treatment options appear promising. These findings were gathered from meta-analyses of approximately 600 patients combined for both outcomes and were robust to subgroup analyses comparing RCTs with observational studies.

With an extensive history and perceived ability to produce compression across fracture sites, TBW remains the most commonly used method of fixation for transverse noncomminuted olecranon fractures.^8,51 However, biomechanical studies have highlighted several weaknesses in TBW, including less compression across fractures relative to precontoured olecranon-specific plates.^21,52,53 Despite biomechanical superiority, the authors found that the functional outcomes (Disabilities of the Arm, Shoulder and Hand score, Mayo Elbow Performance Score, and flexion arc) of TBW and PF are similar in most of the studies. This lack of difference may be the result of studies being underpowered to evaluate functional differences between TBW and PF. However, some authors have shown similar results between TBW and PF in comminuted fractures as well.^34,45,49 Accordingly, there is a lack of consensus in the literature and clinical expert opinion.⁸ Tension-band wiring is favored by some for reasons related to initial implant cost and duration of surgery. One retrospective cohort study using data gathered in the United States reported that TBW was approximately 50% less costly than PF when hospital stay was not accounted for ($6598.36 vs $14,333.46).⁴⁷ Other benefits of TBW may include reduced surgical time.^21,47,49,50

Although the 2 methods lead to similar functional outcomes, this analysis showed that TBW is associated with significantly more complications and more frequent reoperations for hardware removal. Symptomatic (ie, painful or irritating) hardware was the most common complication of both TBW and PF. It would be imprudent to consider symptomatic hardware as trivial because it was also found to be the most common cause of hardware removal. Hardware removal, found to be significantly more likely after TBW, increases patient morbidity through exposure to a second operation and should carry considerable weight in initial implant decision.^54,55 The cost of additional operating room time and hospital admissions associated with removal must also be considered.^31,46 Two recent cost-analyses, 1 cohort study and 1 RCT, both from the United Kingdom, questioned the TBW cost-efficacy dogma; these analyses found that PF was less expensive than TBW ($8374 vs $7812) when all treatment- and admission-related costs were considered.^31,49 However, the difference in implant cost reported between studies was substantial, suggesting cost-efficacy may largely depend on institution-specific implant pricing and protocols.

Patient age should also be taken into consideration when interpreting the current findings. Across the current literature, elderly patients are poorly represented. In the elderly or those with poor bone stock, precontoured olecranon-specific plates have an advantage of increased stability, although screw pullout may still occur.^56,57 Limited soft tissue coverage may also influence application of the current findings to elderly populations. Evidence suggests that nonoperative management or fragment excision may yield equivalent or superior outcomes in elderly, low-demand populations.¹⁰

Several novel treatment methods appear to represent promising alternatives to traditional K-wire TBW, including cable pin system TBW, TBW with cannulated cancellous screws, and the olecranon memory connector. Lu et al¹⁶ suggest that the lower rate of implant removal in patients with TBW plus screw fixation is due to the additional fixation that screws provide, which reduces K-wire migration and is thus less likely to irritate skin and subcutaneous tissues. Potentially, added compression across the fracture site with olecranon screws reduces tension in the figure-of-8 fixation around the triceps tendon, reducing the rates of symptomatic tendonitis. Although all 3 novel methods supported by randomized data appear promising, they have each only been studied in 1 relatively small randomized trial to date. Further evidence comparing these novel methods with PF is required before their use can be advocated for in place of the more common fixation techniques.

This systematic review had several strengths. First, the authors conducted an in-depth literature search of 5 databases in duplicate to ensure all relevant studies were included. The quality of all studies was assessed independently and in duplicate using appropriate quality assessment scales. In addition, the authors conducted subgroup analyses comparing randomized and nonrandomized data to ensure their main outcomes were robust to potential differences in the methodological quality of included studies.

Nevertheless, this systematic review and meta-analysis had several limitations. Studies included in the meta-analysis were of low to moderate quality, with all studies having lack of patient blinding. In addition, the meta-analysis included 3 RCTs. Although the authors were able to support their conclusions with nonrandomized data via subgroup analysis, their primary analyses involved the inherent risks of bias common in observational studies.

Conclusion

Plate fixation yields significantly lower rates of complication as well as hardware removal compared with TBW. This article highlights the need to consider the increased complications associated with TBW before deciding on a method of fixation in adults with olecranon fractures.