Abstract
Purpose
-
To compare the two main surgical approaches to address proximal humerus fractures (PHFs) stratified for Neer fracture types, to demonstrate which approach gives the best result for each fracture type.
Methods
-
A literature search was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines in PubMed, Web of Science, and Cochrane databases up to 4 January 2022. Inclusion criteria were studies comparing open reduction and internal fixation (ORIF) with deltopectoral (DP) approach and minimally invasive plate osteosynthesis (MIPO) with deltosplit (DS) approach of PHFs. Patient’s demographic data, fracture type, Constant–Murley Score (CMS), operation time, blood loss, length of hospital stay, complications, fluoroscopy time, and radiological outcomes were extracted. Results were stratified for each type of Neer fracture.
Results
-
Eleven studies (798 patients) were included in the meta-analysis. No functional difference was found in the CMS between the two groups for each type of Neer (P = n.s.): for PHFs Neer II, the mean CMS was 72.5 (s.e. 5.9) points in the ORIF group and 79.6 (s.e. 2.5) points in the MIPO group; for Neer III, 77.8 (s.e. 2.0) in the ORIF and 76.4 (se 3.0) in the MIPO; and for Neer IV, 70.6 (s.e. 2.7) in the ORIF and 60.9 (s.e. 6.3) in the MIPO. The operation time in the MIPO group was significantly lower than in the ORIF group for both Neer II (P = 0.0461) and Neer III (P = 0.0037) fractures.
Conclusion
-
The MIPO with DS approach demonstrated no significant differences in the results to the ORIF with DP approach for the different Neer fractures in terms of functional results, with a similar outcome, especially for the Neer II and III fracture types. The MIPO technique proved to be as safe and effective as the ORIF approach.
Introduction
Proximal humeral fractures (PHFs) are one of the most common types of fractures, accounting for up to 5% of all fractures, and they are the third most common fracture in the elderly population, increased by more than three times between 1970 and 2002 (1, 2, 3, 4). By 2030, the incidence of PHFs is expected to further grow, due to the ageing population and a more active lifestyle among the elderly (4). The treatment strategy of PHFs depends on different aspects: patient characteristics (age and co-morbidity), injury patterns (fracture type and displacement), and different available options about the surgical technique (5, 6, 7). The majority of PHFs are undisplaced or minimally displaced, thus treated conservatively, while displaced or unstable fractures require surgical treatment to achieve an optimal functional outcome (8, 9). Various surgical techniques, like open reduction and internal fixation (ORIF) with proximal humeral plate, intramedullary nailing, and arthroplasty, have been described in the literature (10, 11). In recent years, the locking plate has become increasingly used for the fixation of PHFs, especially for elderly patients (12, 13, 14).
The optimal surgical access for the locking plate remains controversial, with two main approaches being widely used: ORIF with open deltopectoral (DP) approach and minimally invasive plate osteosynthesis (MIPO) with deltosplit (DS) approach. The first one is the most common (15, 16), being worldwide appreciated for the greater exposure of the fracture rim and the large incision, which can be widened when necessary (17). However, this approach may cause damage to soft tissues and blood vessels, with the risks of injury to the deltoid muscle, increased blood loss, non-union, avascular necrosis (AVN) of the humeral head, and deep infection (13, 18). Given these disadvantages, recently, the MIPO technique (DS approach) has been increasingly used to fix PHFs; the advantages are easy access, shorter duration of surgery, preservation of the soft tissues, better aesthetic result, and earlier exercise of the affected shoulder (18, 19). However, the pitfalls are the risk of damage to the blood supply of the deltoid, and the improper visualization of the axillary nerve, which may suffer iatrogenic injury. In addition, exposure of the medial aspects of the proximal humerus is limited, and anatomical reduction of fractures with more than two parts may be more difficult (20). Accordingly, open surgery is often preferred for more severe cases, while the minimally invasive approach is preferred for less severe fractures. However, up to now, there is still disagreement regarding the best approach in the treatment of PHFs, and there is a lack of conclusive evidence specifically analyzing the results for each fracture pattern (21).
The aim of this systematic review and meta-analysis was to compare the two main approaches for PHFs (ORIF with DP and MIPO with DS) for each type of Neer fracture, documenting which treatment brings a greater benefit in terms of functional scores, operation time, blood loss, length of hospital stay, complications, fluoroscopy time, and radiological outcomes.
Materials and methods
Literature search
A review protocol was created based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (www.prisma-statement.org, accessed on 1 October 2021). The study was registered on PROSPERO (CRD42022312814). A literature search was performed in three bibliographic databases (PubMed, Web of Science, and Wiley Cochrane Library) from inception up to 4 January 2022. The following research terms were used: ‘(proximal humer* OR shoulder OR humer* OR PHF*) AND (Minimally Invasive OR MIPO OR deltoid split approach OR DS approach) AND (ORIF OR deltopectoral OR DP OR Open plating)’.
Inclusion criteria
Inclusion criteria were studies comparing ORIF with DP approach vs MIPO with DS approach for the surgical treatment of PHFs in adults; to be included, the study had to perform the DP vs DS comparison stratifying for each type of Neer fracture. Only studies written in English and reporting a minimum follow-up of 6 months were included. Case reports or case series describing less than five cases and articles in languages other than English were excluded. Pre-clinical studies, ex-vivo studies, and review articles were also excluded.
Data extraction
Two independent reviewers (LMO and AS) screened all the articles on the title and abstract to assess whether they met the inclusion criteria. After the first screening, the articles that met the inclusion criteria were evaluated for full-text eligibility and were excluded if they did not follow the inclusion criteria (Fig. 1). In case of disagreement between the two reviewers (LMO and AS), a third reviewer (PF) was consulted to reach a consensus. Data were independently extracted on a preconceived data extraction form using Excel (Microsoft). The following data were extracted: first author, country, journal, year of publication, type of study, number of patients, type of fracture according to the Neer classification for PHFs (8), Constant–Murley Score (CMS) with at least 6 months of follow-up, operation time, blood loss, length of hospital stay, complications, fluoroscopy time, and radiological outcomes. The CMS is a 100-point scale composed of several questions to assess the function of the shoulder. It evaluates daily pain (0–15 points), range of motion (0–40 points), strength (0–25 points), and ability to carry out daily activities (0–20 points). The total score can be classified as ‘poor’ (0–55 points), ‘fair’ (56–70), ‘good’ (71–85), or ‘excellent’ (86–100) (22). Operation time for both groups, DP and DS, was also reported and measured in minutes. Blood loss during the operation was reported in milliliters (mL). The number and type of complications were also extracted, as well as the rate of non-unions and other radiological findings (malreductions, screw cut-outs, implant loosening, subacromial impingement, and AVN of the humeral head). Total fluoroscopy time during surgery was reported in seconds. After independent data collection, the reviewers compared the extracted data.
PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flowchart.
Citation: EFORT Open Reviews 8, 8; 10.1530/EOR-22-0110
Assessment of risk of bias and quality of evidence
The Downs and Black’s ‘Checklist for Measuring Quality’ was used to evaluate the risk of bias (23). It contains 27 ‘yes’-or-’no’ questions across five sections; it provides a numeric up to 32 points. The five sections include questions about the overall quality of the study (10 items), the ability to generalize findings of the study (3 items), the study bias (7 items), the confounding and selection bias (6 items), and the power of the study (1 item). Assessment of risk of bias and quality of evidence was completed independently for all outcomes by two authors (LMO and AS), and a third author (PF) solved any possible discrepancy.
Statistical analysis
The statistical analysis and the forest plot were carried out according to Neyeloff et al. (24) using Microsoft Excel by an independent professional statistician. The comparisons among the groups were based on the chi-square test and the analysis of variance (25). A statistical test for heterogeneity was first conducted with the Cochran’s Q statistic and I2 metric and was considered the presence of significant heterogeneity with I2 values ≥ 25%. When no heterogeneity was found with I2 < 25%, a fixed effect model was used to estimate the expected values and 95% CIs. Otherwise, a random-effect model was applied, and an I2 metric was evaluated for the random effect to check the correction of heterogeneity. All statistical analysis was carried out using Microsoft Excel 2010.
Results
Details of the included studies
A total of 492 articles were retrieved; after the removal of duplicates and screening on the titles, abstracts, and full-texts, 14 studies (1089 patients) were included in this systematic review. Eleven studies reported patient gender, 326 men (34.8%) and 611 women (65.2%). The mean age was 62.8 years in the DP group and 63.0 years in the DS group. The mean follow-up time was 20.2 months in the DP group and 18.7 months in the DS group (Table 1 for further details). Eleven studies presented relevant data which could be included in the meta-analysis (Fig. 1). The meta-analysis included 798 patients, 447 in the DP group and 351 in the DS group. According to the Neer classification for PHFs, a Neer II fracture was diagnosed in 89 patients in the ORIF group and 102 patients in the MIPO group; a Neer III fracture was diagnosed in 325 patients in the ORIF group and 217 in the MIPO group; and a Neer IV fracture was found in 33 patients in the ORIF group and 32 patients in the MIPO group.
Details of the included studies.
| Reference | Country | Study type | Patients, n | Age (years) | Sex | Neer fracture type | Studies in the SR | Studies in the meta-analysis | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DP | DS | M/F | II DP | II DS | III DP | III DS | IV DP | IV DS | CMS | OpTime | |||||
| Borer et al. (26) | Switzerland | PS | 62 | 62.0 | 67.0 | 16/46 | 7 | 12 | 11 | 20 | 5 | 7 | ✓ | ✓ | |
| Buchmann et al. (27) | Switzerland | RS | 198 | 64.8 | 63.9 | 75/123 | 0 | 0 | 149 | 49 | 0 | 0 | ✓ | | ✓ |
| Buecking et al. (28) | Germany | RCT | 120 | 67.0 | 69.0 | 28/92 | 15 | 15 | 45 | 45 | 0 | 0 | ✓ | ✓ | ✓ |
| Chiewchantanakit et al. (5) | Thailand | RS | 28 | 62.2 | 51.9 | 12/16 | 6 | 8 | 10 | 4 | 0 | 0 | ✓ | | ✓ |
| Hepp et al. (29) | Germany | RS | 83 | 65.0 | 65.0 | 19/64 | 6 | 12 | 31 | 22 | 7 | 5 | ✓ | ✓ | |
| Kim et al. (30) | South Korea | RS | 36 | 52.6 | 58.7 | N/A | 17 | 19 | 0 | 0 | 0 | 0 | ✓ | ✓ | ✓ |
| Lin et al. (31) | China | RS | 86 | 61.0 | 63.0 | 28/58 | 9 | 10 | 22 | 24 | 12 | 9 | ✓ | ✓ | |
| Liu et al. (32) | China | RS | 91 | 61.7 | 60.2 | 42/49 | 15 | 18 | 18 | 25 | 9 | 6 | ✓ | | |
| Rouleau et al. (33) | Canada | RCT | 85 | 62.0 | 63.0 | 19/66 | 21 | 20 | 14 | 20 | 2 | 0 | ✓ | | |
| Sohn et al. (34) | South Korea | RCT | 90 | 62.6 | 61.0 | N/A | 18 | 17 | 23 | 21 | 4 | 7 | ✓ | ✓ | ✓ |
| Vijayvargiya et al. (35) | India | PS | 26 | 46.0 | 46.0 | N/A | 3 | 2 | 5 | 7 | 5 | 4 | ✓ | ✓ | |
| Wang et al. (36) | China | RS | 115 | 62.0 | 62.0 | 52/63 | 16 | 18 | 21 | 30 | 14 | 16 | ✓ | | |
| Zhang et al. (37) | China | RS | 33 | 62.6 | 66.2 | 14/19 | 0 | 0 | 20 | 13 | 0 | 0 | ✓ | ✓ | |
| Zhao et al. (38) | China | RCT | 36 | 63.6 | 64.3 | 21/15 | 8 | 7 | 9 | 12 | 0 | 0 | ✓ | ✓ | ✓ |
CMS, Constant–Murley Score; DP, deltopectoral; DS, deltosplit; F, female; M, male; N/A, not available; Op, operation; PS, prospective study; RCT, randomized controlled trial; RS, retrospective study; SR, systematic review.
Systematic review results
All the studies included in this systematic review reported the complications that occurred in the DP and DS groups (5, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38). Overall, there were no differences in the number of complications: 121 in the DP group (28.4%) and 105 in the DS group (29.9%) (P = n.s.). The detailed type and number of complications that occurred in both groups are reported in Table 2. A significantly higher incidence of screw cut-outs/implant loosening was documented in the MIPO group (n = 36, 10.3%) compared to the ORIF group (n = 25, 5.6%) (P = 0.01). However, only five studies (27, 30, 32, 34, 37) took into account the complication rate for Neer PHFs type subgroups. Complications for Neer II fractures were described by two studies: out of 71 patients, the complication rate was 8.6% (3 out of 35 patients) in the DP group and 2.8% (1 out of 36 patients) in the DS group (30, 34). For Neer III fractures, four studies reported a complication rate of 10.5% in the DP group (22 out of 210 patients) and 11.1% (12 out of 108 patients) in the DS group (27, 32, 34, 37). For Neer IV fractures, two studies reported the complication rate, but the sample size was very small, thus making the results poorly reliable: the complication rate was 69.2% (9 out of 13 patients) in the DP group and 53.8% (7 out of 13 patients) in the DS group (32, 34). Overall, there were 43 reinterventions in the DP group (9.6%) and 49 in the DS group (13.9%).
Number and type of complications. Data are presented as n (%).
| Type of complication | Patients | P-value* | |
|---|---|---|---|
| DP (n = 447) | DS (n = 351) | ||
| AVN | 11 (2.5%) | 17 (4.8%) | 0.07 |
| Subacromial impingement | 12 (2.7%) | 15 (4.3%) | 0.22 |
| Screw cut-outs/implant loosening | 25 (5.6%) | 36 (10.3%) | 0.01 |
| Infection | 12 (2.7%) | 4 (1.1%) | 0.12 |
| Malreduction | 23 (5.1%) | 16 (4.6%) | 0.7 |
| Non-union | 29 (6.5%) | 5 (1.4 %) | 0.0004 |
| Stiffness | 8 (1.8%) | 4 (1.1%) | 0.45 |
| Delayed union | 1 (0.2%) | 2 (0.6%) | 0.43 |
| Implant failure | 1 (0.2%) | 3 (0.9%) | 0.21 |
| Axillary nerve injury | 3 (0.7%) | 1 (0.3%) | 0.44 |
| Radial nerve injury | 2 (0.4%) | 0 | 0.21 |
| Persistent pain | 0 | 1 (0.3%) | 0.26 |
| Plate broken | 0 | 1 (0.3%) | 0.26 |
| Total | 121 (28.4%) | 105 (29.9%) | 0.38 |
*P-values compare DP and DS and statistically significant results are typed in bold.
AVN, avascular necrosis of the humeral head; DP, deltopectoral; DS, deltosplit.
Radiological follow-ups were documented in all the included studies. Malreduction was observed in 23 patients in the DP group (5.2%) and 16 patients in the DS group (4.6%). There were 29 non-unions in the DP group (6.5%) and 5 non-unions in the DS group (1.4%) (P = 0.0004). In addition, there were 11 AVN of the humeral head in the DP group (2.5%) and 17 in the DS group (4.8%) (P = 0.07) (Table 2). Unfortunately, no Neer subgroup analysis was performed to understand if there were differences between the two approaches based on the Neer fracture type.
Five studies reported blood loss during surgery for both DP and DS groups (5, 31, 32, 33, 38). However, a meta-analysis was not feasible since only two studies assessed the blood loss for Neer II and Neer III fractures (5, 38). The mean blood loss for Neer II PHFs was 127.6 mL in the DP group and 114.5 mL in the DS group. For Neer III fractures, the blood loss was 131.2 mL in the DP group and 119.5 mL in the DS group. Two studies reported the fluoroscopy time for Neer II PHFs and the mean radiation exposure time was 34.6 s in the DP group and 69.2 s in the DS group (28, 30). Two studies reported the fluoroscopy time for Neer III PHFs and for the DP group was 148.5 s and 154.9 s in the DS group (27, 28). None of the studies reported fluoroscopy time for Neer IV fractures. Four studies reported the length of hospital stay (5, 26, 28, 37), but only one study described it for Neer II fractures and two studies for Neer III fractures (5, 37). For Neer II fractures, the mean hospital stay was 11.3 days in the DP group and 6.2 days in the DS group. For Neer III fractures, the mean hospital stay was 7.5 days in the DP group and 7.7 days in the DS group.
Meta-analysis on patient-reported outcomes
Nine studies reported the CMS at 12 months of follow-up: for all the different types of Neer fractures, no statistical difference between the two approaches was found (P = n.s.). In particular, eight studies reported the CMS for Neer II fractures: the mean was 72.5 points (s.e.: 5.9, 95% CI: 61.0–84.0) in the DP group and 79.6 points (s.e.: 2.5, 95% CI: 74.7–84.4) in the DS group (26, 28, 29, 31, 34, 35, 38), with a difference of 7.0 points (95% CI: 8.0–22.0) between the two groups (Fig. 2). Eight studies reported the CMS for Neer III fractures: the mean was 77.8 points (s.e.: 2.0, 95% CI: 73.8–81.7) in the DP group and 76.4 points (s.e.: 3.0, 95% CI: 70.5–82.2) in the DS group (26, 28, 29, 31, 34, 35, 37, 38), with a difference of 1.0 point (95% CI: 5.0–3.1) between the two groups (Fig. 3). Five studies reported the CMS for Neer IV fractures, and the mean was 70.6 (s.e.: 2.7, 95% CI: 65.4–75.9) in the DP group and 60.9 (s.e.: 6.3, 95% CI: 48.6–73.2) in the DS group (26, 29, 31, 34, 35), with a difference of 8.1 points (95% CI: −24.5 to 8.2) between the two groups (Table 3) (Fig. 4).
Forest plot of weighted mean difference and 95% confidence intervals (CI) Constant–Murley Score (CMS) for Neer II. Std, standard.
Citation: EFORT Open Reviews 8, 8; 10.1530/EOR-22-0110
Forest plot of weighted mean difference and 95% confidence intervals (CI) for Constant–Murley Score (CMS) for Neer III. Std, standard.
Citation: EFORT Open Reviews 8, 8; 10.1530/EOR-22-0110
Forest plot of weighted mean difference and 95% confidence intervals (CI) for Constant–Murley Score (CMS) for Neer IV. Std, standard.
Citation: EFORT Open Reviews 8, 8; 10.1530/EOR-22-0110
Meta-analysis results.
| Outcome/Approach | Number of | Results | |||
|---|---|---|---|---|---|
| Studies | Patients | Mean | 95% CI | P-value† | |
| Constant–Murley Score | |||||
| Neer II | 0.26 | ||||
| DP | 8 | 83 | 72.5 | 61.0–84.0 | |
| DS | 8 | 94 | 79.6 | 74.7–84.4 | |
| Neer III | 0.35 | ||||
| DP | 8 | 166 | 77.8 | 73.8–81.7 | |
| DS | 8 | 164 | 76.4 | 70.5–82.2 | |
| Neer IV | 0.24 | ||||
| DP | 5 | 33 | 70.6 | 65.4–75.9 | |
| DS | 5 | 32 | 60.9 | 48.6–73.2 | |
| Operation time* | |||||
| Neer II | 0.046 | ||||
| DP | 5 | 64 | 91.0 | 73.6–108.4 | |
| DS | 5 | 66 | 73.2 | 63.1–83.3 | |
| Neer III | 0.0037 | ||||
| DP | 5 | 236 | 95.8 | 76.1–115.5 | |
| DS | 5 | 131 | 72.4 | 61.8–83.0 | |
*The operation time was reported in minutes; †Statistically significant values are in bold.
CI, confidence interval; DP, deltopectoral; DS, deltosplit.
Meta-analysis on operation time
A total of 12 studies reported the operation time for the ORIF and MIPO groups. However, six of them were not included in the meta-analysis because they reported the operation time without distinguishing for each type of Neer’s fracture. Among the six studies included in the meta-analysis, five reported the mean operation time for Neer II fractures and five for Neer III fractures. For Neer IV fractures, a meta-analysis was not feasible since only two studies provided the operation time. The operation time in the MIPO group was significantly lower than in the ORIF group for both Neer II (P = 0.0461) and Neer III (P = 0.0037) fractures. For Neer II fractures, the mean operation time was 91.0 min (s.e.: 8.9, 95% CI: 73.6–108.4) in the DP group and 73.2 min (s.e.: 5.2, 95% CI: 63.1–83.3) in the DS group (5, 28, 30, 34, 38) (Fig. 5). For Neer III fractures, the mean operation time was 95.8 min (s.e.: 10.1, 95% CI: 76.1–115.5) in the DP group and 72.4 min (s.e.: 5.4, 95% CI: 42.9–67.8) in the DS group (Table 3) (5, 27, 28, 34, 38) (Fig. 6).
Forest plot of weighted mean difference and 95% confidence intervals (CI) for operation time for Neer II. Std, standard.
Citation: EFORT Open Reviews 8, 8; 10.1530/EOR-22-0110
Forest plot of weighted mean difference and 95% confidence intervals (CI) for operation time for Neer III.
Citation: EFORT Open Reviews 8, 8; 10.1530/EOR-22-0110
Risk of bias
The Downs and Black’s tool for assessing the risk of bias gives each study an excellent ranking for scores ≥ 26, good for scores from 20 to 25, fair for scores between 15 and 19, and poor for scores ≤ 14 points. According to these criteria, among the included studies, 0 studies were classified as poor, 1 fair, 11 good, and 2 excellent (Fig. 7). Mostly, the factors reducing the quality of the studies were confounders, lack of proper blinding, and low statistical power of some studies.
Downs and Black’s tool for assessing the risk of bias.
Citation: EFORT Open Reviews 8, 8; 10.1530/EOR-22-0110
Discussion
The main findings of this systematic review and meta-analysis were that MIPO with DS approach provides comparable shoulder function with respect to ORIF with DP approach, with a similar outcome especially for the Neer II and III fracture types, while presenting advantages in terms of lower operation time.
Despite the continuous increase in the number of PHFs (4), only a few reports on the clinical outcomes of the DP and DS surgical approaches have been published so far, especially when considering the need to account for the main confounder: the type of fracture. When evaluating the effectiveness of a surgical approach, postoperative shoulder function is one of the elements that help analyse the ORIF with DP and the MIPO with DS approaches. Different scores have been used to assess shoulder function; the CMS is a physician-assessed score largely appreciated by the literature to record patient outcomes after shoulder surgery, thanks to its reliability and completeness (28). The present study demonstrated comparable CMS results between these two approaches in the fixation of all types of fractures. In Neer II PHFs, the DS approach group obtained a CMS value higher of 7.0 points than the DP approach group. This ratio reversed in the treatment of Neer III PHFs in favour of the DP group, with a mean difference of 1.0 point. This could have been expected since an open approach could better address a more complex fracture. However, it must be underlined that both approaches demonstrated for both fracture types values of CMS that were considered as ‘good’, according to the definition provided by Constant et al. (22). In Neer IV PHFs, the mean difference between the approaches was 8.1 points in favour of the DP group (P = n.s.); in both treatment groups, a ‘fair’ CMS value was found, which could probably be caused by the higher fracture complexity. Because the DS approach is less invasive than the DP, many authors described through the years the possibility of a superior shoulder function in Neer II PHFs treated with this technique, as the soft tissues suffer a relatively small injury from the two-part fracture, and the DS approach helps to preserve them (39, 40). Conversely, in the Neer IV PHFs, it was thought that the MIPO treatment, due to less exposure of the fracture line, did not allow optimal fixation and a functional outcome as high as in the ORIF group (41, 42). The current study questions the significance of this possible difference in the treatment outcome, demonstrating that the results between the two groups were not statistically different for each fracture type even in more complex fractures, where an ORIF approach has been generally considered to have advantages (20, 43). The results of the current study can be explained by the fact that in the MIPO approach, the undamaged muscles and soft tissues can help to successfully restore the function of the shoulder joint, and this result confirms that ‘shoulder surgery is a surgery of the soft parts’, as C.S. Neer said (44). Previous literature analyses suggested similar results for the two approaches (43, 45, 46) but, since no study normalized the data for each type of fracture and rather analysed them as a whole, they were affected by an assessment bias. The results of this meta-analysis were able to demonstrate the potential of the two approaches for the different types of fractures, shedding new light on this field which helps in defining pros, cons, and indications on the most suitable approach for PHFs.
Another important aspect evaluated in the present study was the radiological outcome. The ORIF group showed a statistically significant higher rate of non-unions compared to the MIPO group (6.5 vs 1.4%, respectively), while no differences could be documented in the malreduction rate. Unfortunately, a further analysis of the radiological complications stratified for each Neer fracture type was not feasible due to the lack of data. According to the available literature, MIPO with DS approach is believed to provide a better fracture union process compared to the more invasive ORIF technique (21, 30). Some authors suggested that the minor soft tissue trauma with the MIPO approach, together with the preservation of the periosteal circulation, may accelerate fracture healing (43, 46). On the other hand, more screw cut-outs and implant loosening were documented in the MIPO group (10.3 vs 5.6%). The results of this review are in line with the previous literature, although further studies are needed to compare the radiological outcomes of the two techniques for each Neer fracture type. Finally, there were more AVN of the humeral head in the DS group compared to the DP group. Although without reaching statistical significance, a tendency was noted (P = 0.07). This result is in line with a previous meta-analysis (21). Unfortunately, a sub-analysis on the Neer fracture type was not feasible for this type of complication due to the lack of data. AVN is a well-known complication of PHFs (47) and heterogeneous incidence rates are reported across the literature for both approaches (48, 49). These controversial findings should be addressed by future studies stratifying Neer fracture types, in order to point out the potential impact of the surgical technique on the development of this severe complication.
The operative times of the two approaches were analysed separately for each fracture type. This meta-analysis found a statistically lower operative time in the MIPO group compared to the ORIF group for the fixation of Neer type II and III PHFs; for type IV PHFs the meta-analysis was not feasible due to the lack of sufficient data. However, the study by Sohn et al. evaluated the operative time and reported a shorter duration in patients operated with MIPO with DS approach also for Neer IV fractures (34). This difference in operation time is explained by the fact that the DS approach can provide a sufficient view of the fracture line and of the position of the plate, even in more complex fractures, while requiring only minimal soft tissue dissection (50). Moreover, the position of this dissection guarantees that all the vascular and nervous structures are safely distant, thus it can be performed faster (42, 50). Although there is preclinical evidence of an increased risk of axillary nerve damage with the DS approach (51), a precise incision, aware of these risks, prevents from such complications (52), and among the included studies, only one axillary nerve injury was documented in the DS group (study by Lin et al. on 86 patients (31)), while three occurred in the DP group (27). Owing to the aforementioned reasons, the shorter operation time of MIPO is certainly an advantage over ORIF, especially considering that patients with Neer IV PHFs are often polytraumatized, requiring more surgeries (33, 34) and that the final functional outcome is similar. Finally, Zhao et al. (38) confirmed that even in patients with a BMI > 26 undergoing MIPO treatment the surgery duration was statistically shorter to the one of the ORIF group, for both Neer II and Neer III fractures.
MIPO with DS approach has also some disadvantages, as it requires repetitive use of fluoroscopy to obtain an acceptable reduction. While the literature data were not suitable for a meta-analysis, the available findings point in this direction. Kim et al. (30) in their clinical study showed a statistically higher radiation exposure for the MIPO group than the ORIF group for Neer II PHF. Buecking et al. (28) in their study of Neer III and Neer IV PHFs also suggested a longer mean exposure time for the MIPO group. The longer radiation exposure time is a major disadvantage of the MIPO technique compared to the ORIF technique, and surgeons should be cautious about the radiation hazard during the operation, although further studies specifically taking into account this aspect should be conducted (21).
Other aspects can differ between the two approaches. Previous reviews showed that MIPO treatment for PHFs results in significantly less blood loss during surgery than ORIF treatment (31, 43, 45, 46), but this conclusion was weakened by the fact that all previous literature did not distinguish PHFs by Neer type, thus providing a result that does not take into account the important differences between each fracture type in terms of complexity and consequences. Up to now, there are only two clinical studies, by Chiewchantanakit et al. and Zhao et al., that report blood loss separately for Neer II and Neer III PHFs (5, 38), and no study focusing on Neer IV. Chiewchantanakit et al. showed that the mean blood loss in Neer II was the same between the two treatments, whereas for Neer III, the MIPO group reported a lower mean blood loss than the ORIF group, but no statistical correlation was found (5). Zhao et al. reported similar results for both Neer II and Neer III, and they showed that patients with a BMI > 26 operated in the MIPO group tended to have lower blood loss than the ORIF group for both Neer II and Neer III fractures (38). No meta-analysis of these data was feasible, and future studies should evaluate this aspect due to its clinical relevance, especially in older and complex patients.
Another important factor to consider when comparing ORIF and MIPO is the hospital length of stay, as the MIPO technique minimizes the incision and avoids damage to the deltoid muscle, causing a smaller surgical wound and reducing post-operative pain. In this systematic review, only two articles reported the average length of stay for each type of Neer analysed. Kim et al. retrieved a statically shorter hospitalization time in the MIPO group for Neer II PHFs (30), whereas Neer III PHFs showed a comparable mean hospitalization time, as also described by Zhang et al. (37). Finally, when different surgical approaches are analysed, a crucial aspect to be considered is the complication rate. Overall, the number of complications was similar in both groups, accounting for 28.4% in the DP group and 29.9% in the DS group (P = n.s.), although with some differences, with the most frequently observed being non-unions in the DP group (6.5%) and screw cut-outs/implant loosening in the DS group (10.3%). Unfortunately, only five studies (27, 30, 32, 34, 37) compared the complication rate between DP and DS approaches for the different types of PHFs, but the data were not meta-analysed due to the heterogeneity of the results. However, an important finding was still retrieved from the systematic review: the fewer complications of the DS approach in Neer type IV when compared to the DP approach reported in the small cohorts documented by Sohn et al. (34). It is important that future comparative studies address this question because while there is a common belief among surgeons that for Neer IV a MIPO approach leads to a higher complication rate because of less exposure of the fracture rim, the data provided by the literature so far, although limited to few cases, seem to point to another direction (27, 33). Finally, similar rates of reinterventions were documented in both groups, although no sufficient data were available for a sub-analysis on the fracture pattern, nor for the incidence of conversion to total shoulder arthroplasty.
Even if this meta-analysis and systematic review allowed to draw relevant conclusions, there are still some limitations and biases. The first limitation is the important heterogeneity of the data, and the small number of studies included. Moreover, it was possible to analyse separately based on the fracture Neer type only CMS and operation time, since most of the studies did not report the stratified data: therefore, a cautious interpretation of the results is recommended, especially for Neer IV fractures, for which the sample size was very limited. Secondly, different surgeons and different hospitals had different surgical protocols, approaches, materials (such as calcar screws for comminuted fractures) and possibly also indications, which may cause bias. More rigorous designs and large RCTs are needed to further confirm the data retrieved in this meta-analysis, especially for Neer IV fractures. Despite these limitations, this is the first study to compare the MIPO with DS and ORIF with DP approaches in PHFs dividing their outcomes by Neer fracture type, questioning common surgical beliefs, and shedding new light on the potential of the MIPO DS approach to successfully address also the more complex Neer PHFs.
Conclusion
According to this systematic review, the MIPO with DS approach provides comparable shoulder function with respect to ORIF with DP approach, with a similar outcome especially for the Neer II and III fracture types, while showing a shorter operation time than the ORIF group. The MIPO technique proved to be as safe and effective as the ORIF approach. While the available evidence suggests the MIPO technique as a valid option to address each type of PHF, more studies are needed to confirm these findings by comparing the two approaches for each Neer fracture type.
ICMJE conflict of interest statement
The authors decalre that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding Statement
This study did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.
References
- 1.↑
Roux A, Decroocq L, El Batti S, Bonnevialle N, Moineau G, Trojani C, Boileau P, & de Peretti F. Epidemiology of proximal humerus fractures managed in a trauma center. Orthopaedics and Traumatology, Surgery and Research 2012 98 715–719. (https://doi.org/10.1016/j.otsr.2012.05.013)
- 2.↑
Baron JA, Karagas M, Barrett J, Kniffin W, Malenka D, Mayor M, & Keller RB. Basic epidemiology of fractures of the upper and lower limb among Americans over 65 years of age. Epidemiology 1996 7 612–618. (https://doi.org/10.1097/00001648-199611000-00008)
- 3.↑
Lippuner K, Popp AW, Schwab P, Gitlin M, Schaufler T, Senn C, & Perrelet R. Fracture hospitalizations between years 2000 and 2007 in Switzerland: a trend analysis. Osteoporosis International 2011 22 2487–2497. (https://doi.org/10.1007/s00198-010-1487-8)
- 4.↑
Kannus P, Palvanen M, Niemi S, Parkkari J, Järvinen M, & Vuori I. Osteoporotic fractures of the proximal humerus in elderly Finnish persons: sharp increase in 1970–1998 and alarming projections for the new millennium. Acta Orthopaedica Scandinavica 2000 71 465–470. (https://doi.org/10.1080/000164700317381144)
- 5.↑
Chiewchantanakit S, & Tangsripong P. Locking plate fixation of proximal humeral fracture: minimally invasive vs. standard delto-pectoral approach. Journal of the Medical Association of Thailand = Chotmaihet Thangphaet 2015 98 196–200.
- 6.↑
Berkes MB, Little MT, & Lorich DG. Open reduction internal fixation of proximal humerus fractures. Current Reviews in Musculoskeletal Medicine 2013 6 47–56. (https://doi.org/10.1007/s12178-012-9150-y)
- 7.↑
Marin R, Feltri P, Ferraro S, Ippolito G, Campopiano G, Previtali D, Filardo G, Marbach F, De Marinis G, Candrian C, et al.Impact of tuberosity treatment in reverse shoulder arthroplasty after proximal humeral fractures: a multicentre study. Journal of Orthopaedic Science 2022. (https://doi.org/10.1016/j.jos.2022.03.008)
- 8.↑
Neer CS 2nd. Displaced proximal humeral fractures. I. Classification and evaluation. Journal of Bone and Joint Surgery. American Volume 1970 52 1077–1089. (https://doi.org/10.2106/00004623-197052060-00001)
- 9.↑
Neer CS 2nd. Displaced proximal humeral fractures. II. Treatment of three-part and four-part displacement. Journal of Bone and Joint Surgery. American Volume 1970 52 1090–1103.
- 10.↑
Lungershausen W, Bach O, & Lorenz CO. Locking plate osteosynthesis for fractures of the proximal humerus. Zentralblatt fur Chirurgie 2003 128 28–33. (https://doi.org/10.1055/s-2003-37368)
- 11.↑
Resch H, Povacz P, Fröhlich R, & Wambacher M. Percutaneous fixation of three- and four-part fractures of the proximal humerus. Journal of Bone and Joint Surgery. British Volume 1997 79 295–300. (https://doi.org/10.1302/0301-620x.79b2.6958)
- 12.↑
Moonot P, Ashwood N, & Hamlet M. Early results for treatment of three- and four-part fractures of the proximal humerus using the PHILOS plate system. Journal of Bone and Joint Surgery. British Volume 2007 89 1206–1209. (https://doi.org/10.1302/0301-620X.89B9.18528)
- 13.↑
Oldrini LM, Feltri P, Albanese J, Marbach F, Filardo G, & Candrian C. PHILOS synthesis for proximal humerus fractures has high complications and reintervention rates: a systematic review and meta-analysis. Life 2022 12 311. (https://doi.org/10.3390/life12020311)
- 14.↑
Oldrini LM, Feltri P, Albanese J, Marbach F, Filardo G, & Candrian C. Reply to Sidiropoulos, K.; Tsikopoulos, K. Comment on "Oldrini et al. PHILOS synthesis for proximal humerus fractures has high complications and reintervention rates: a systematic review and meta-analysis. Life 2022 12 1282. (https://doi.org/10.3390/life12081282)
- 15.↑
Khatib O, Onyekwelu I, & Zuckerman JD. The incidence of proximal humeral fractures in New York State from 1990 through 2010 with an emphasis on operative management in patients aged 65 years or older. Journal of Shoulder and Elbow Surgery 2014 23 1356–1362. (https://doi.org/10.1016/j.jse.2013.12.034)
- 16.↑
Sumrein BO, Huttunen TT, Launonen AP, Berg HE, Felländer-Tsai L, & Mattila VM. Proximal humeral fractures in Sweden-a registry-based study. Osteoporosis International 2017 28 901–907. (https://doi.org/10.1007/s00198-016-3808-z)
- 17.↑
Brunner F, Sommer C, Bahrs C, Heuwinkel R, Hafner C, Rillmann P, Kohut G, Ekelund A, Muller M, Audigé L, et al.Open reduction and internal fixation of proximal humerus fractures using a proximal humeral locked plate: a prospective multicenter analysis. Journal of Orthopaedic Trauma 2009 23 163–172. (https://doi.org/10.1097/BOT.0b013e3181920e5b)
- 18.↑
Bockmann B, Buecking B, Franz D, Zettl R, Ruchholtz S, & Mohr J. Mid-term results of a less-invasive locking plate fixation method for proximal humeral fractures: a prospective observational study. BMC Musculoskeletal Disorders 2015 16 160. (https://doi.org/10.1186/s12891-015-0618-y)
- 19.↑
Gallo RA, Zeiders GJ, & Altman GT. Two-incision technique for treatment of complex proximal humerus fractures. Journal of Orthopaedic Trauma 2005 19 734–740. (https://doi.org/10.1097/01.bot.0000174708.88108.da)
- 20.↑
Gavaskar AS, Chowdary N, & Abraham S. Complex proximal humerus fractures treated with locked plating utilizing an extended deltoid split approach with a shoulder strap incision. Journal of Orthopaedic Trauma 2013 27 73–76. (https://doi.org/10.1097/BOT.0b013e31825cf545)
- 21.↑
Li F, Liu X, Wang F, Gu Z, Tao Q, Yao C, Luo X, & Nie T. Comparison between minimally invasive plate osteosynthesis and open reduction-internal fixation for proximal humeral fractures: a meta-analysis based on 1050 individuals. BMC Musculoskeletal Disorders 2019 20 550. (https://doi.org/10.1186/s12891-019-2936-y)
- 22.↑
Constant CR, Gerber C, Emery RJ, Søjbjerg JO, Gohlke F, & Boileau P. A review of the Constant score: modifications and guidelines for its use. Journal of Shoulder and Elbow Surgery 2008 17 355–361. (https://doi.org/10.1016/j.jse.2007.06.022)
- 23.↑
Downs SH, & Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. Journal of Epidemiology and Community Health 1998 52 377–384. (https://doi.org/10.1136/jech.52.6.377)
- 24.↑
Neyeloff JL, Fuchs SC, & Moreira LB. Meta-analyses and Forest plots using a Microsoft Excel spreadsheet: step-by-step guide focusing on descriptive data analysis. BMC Research Notes 2012 5 52. (https://doi.org/10.1186/1756-0500-5-52)
- 25.↑
Borenstein M, Hedges LV, Higgins JPT, & Rothstein HR. A basic introduction to fixed-effect and random-effects models for meta-analysis. Research Synthesis Methods 2010 1 97–111. (https://doi.org/10.1002/jrsm.12)
- 26.↑
Borer J, Schwarz J, Potthast S, Jakob M, Lenzlinger P, Zingg U, & Babians A. Mid-term results of minimally invasive deltoid-split versus standard open deltopectoral approach for PHILOS™ (proximal humeral internal locking system) osteosynthesis in proximal humeral fractures. European Journal of Trauma and Emergency Surgery 2020 46 825–834. (https://doi.org/10.1007/s00068-019-01076-7)
- 27.↑
Buchmann L, van Lieshout EMM, Zeelenberg M, den Hartog D, Pfeifer R, Allemann F, Pape HC, & Halvachizadeh S. Proximal humerus fractures (PHFs): comparison of functional outcome 1 year after minimally invasive plate osteosynthesis (MIPO) versus open reduction internal fixation (ORIF). European Journal of Trauma and Emergency Surgery 2022 48 4553–4558. (https://doi.org/10.1007/s00068-021-01733-w)
- 28.↑
Buecking B, Mohr J, Bockmann B, Zettl R, & Ruchholtz S. Deltoid-split or deltopectoral approaches for the treatment of displaced proximal humeral fractures? Clinical Orthopaedics and Related Research 2014 472 1576–1585. (https://doi.org/10.1007/s11999-013-3415-7)
- 29.↑
Hepp P, Theopold J, Voigt C, Engel T, Josten C, & Lill H. The surgical approach for locking plate osteosynthesis of displaced proximal humeral fractures influences the functional outcome. Journal of Shoulder and Elbow Surgery 2008 17 21–28. (https://doi.org/10.1016/j.jse.2007.03.029)
- 30.↑
Kim YG, Park KH, Kim JW, Oh JK, Yoon JP, Kim HJ, et al.Is minimally invasive plate osteosynthesis superior to open plating for fixation of two-part fracture of the proximal humerus? Journal of Orthopaedic Surgery (Hong Kong) 2019 27 2309499019836156.
- 31.↑
Lin T, Xiao B, Ma X, Fu D, & Yang S. Minimally invasive plate osteosynthesis with a locking compression plate is superior to open reduction and internal fixation in the management of the proximal humerus fractures. BMC Musculoskeletal Disorders 2014 15 206. (https://doi.org/10.1186/1471-2474-15-206)
- 32.↑
Liu K, Liu PC, Liu R, & Wu X. Advantage of minimally invasive lateral approach relative to conventional deltopectoral approach for treatment of proximal humerus fractures. Medical Science Monitor 2015 21 496–504. (https://doi.org/10.12659/MSM.893323)
- 33.↑
Rouleau DM, Balg F, Benoit B, Leduc S, Malo M, Vézina F, & Laflamme GY. Deltopectoral vs. deltoid split approach for proximal humerus fracture fixation with locking plate: a prospective RAndomized study (HURA). Journal of Shoulder and Elbow Surgery 2020 29 2190–2199. (https://doi.org/10.1016/j.jse.2020.06.020)
- 34.↑
Sohn HS, Jeon YS, Lee J, & Shin SJ. Clinical comparison between open plating and minimally invasive plate osteosynthesis for displaced proximal humeral fractures: a prospective randomized controlled trial. Injury 2017 48 1175–1182. (https://doi.org/10.1016/j.injury.2017.03.027)
- 35.↑
Vijayvargiya M, Pathak A, & Gaur S. Outcome analysis of locking plate fixation in proximal humerus fracture. Journal of Clinical and Diagnostic Research 2016 10 RC01–RC05. (https://doi.org/10.7860/JCDR/2016/18122.8281)
- 36.↑
Wang JQ, Lin CC, Zhao YM, Jiang BJ, & Huang XJ. Comparison between minimally invasive deltoid-split and extended deltoid-split approach for proximal humeral fractures: a case-control study. BMC Musculoskeletal Disorders 2020 21 406. (https://doi.org/10.1186/s12891-020-03417-9)
- 37.↑
Zhang Z, Zhang G, Peng Y, Wang X, Guo H, Zhang W, Tang P, & Zhang L. Modified minimally invasive approach and intra-osseous portal for three-part proximal humeral fractures: a comparative study. Journal of Orthopaedic Surgery and Research 2018 13 24. (https://doi.org/10.1186/s13018-017-0701-1)
- 38.↑
Zhao L, Yang P, Zhu L, & Chen AM. Minimal invasive percutaneous plate osteosynthesis (MIPPO) through deltoid-pectoralis approach for the treatment of elderly proximal humeral fractures. BMC Musculoskeletal Disorders 2017 18 187. (https://doi.org/10.1186/s12891-017-1538-9)
- 39.↑
Acklin YP, Jenni R, Walliser M, & Sommer C. Minimal invasive PHILOS(®)-plate osteosynthesis in proximal humeral fractures. European Journal of Trauma and Emergency Surgery 2009 35 35–39. (https://doi.org/10.1007/s00068-008-7154-5)
- 40.↑
Frima H, Michelitsch C, Beks RB, Houwert RM, Acklin YP, & Sommer C. Long-term follow-up after MIPO Philos plating for proximal humerus fractures. Archives of Orthopaedic and Trauma Surgery 2019 139 203–209. (https://doi.org/10.1007/s00402-018-3063-1)
- 41.↑
Oh HK, Cho DY, Choo SK, Park JW, Park KC, & Lee JI. Lessons learned from treating patients with unstable multifragmentary fractures of the proximal humerus by minimal invasive plate osteosynthesis. Archives of Orthopaedic and Trauma Surgery 2015 135 235–242. (https://doi.org/10.1007/s00402-014-2138-x)
- 42.↑
Ortmaier R, Filzmaier V, Hitzl W, Bogner R, Neubauer T, Resch H, & Auffarth A. Comparison between minimally invasive, percutaneous osteosynthesis and locking plate osteosynthesis in 3-and 4-part proximal humerus fractures. BMC Musculoskeletal Disorders 2015 16 297. (https://doi.org/10.1186/s12891-015-0770-4)
- 43.↑
Zang JC, Du JJ, Li C, Wang JB, & Ma XL. Comparison between minimally invasive plate osteosynthesis and open plating for proximal humeral fractures: a meta-analysis. Journal of Comparative Effectiveness Research 2018 7 1001–1008. (https://doi.org/10.2217/cer-2018-0042)
- 44.↑
Neer CS. The classic: articular replacement for the humeral head. 1955. Clinical Orthopaedics and Related Research 2011 469 2409–2421. (https://doi.org/10.1007/s11999-011-1944-5)
- 45.↑
Zhao W, Zhang Y, Johansson D, Chen X, Zheng F, & Li L. Comparison of minimally invasive percutaneous plate osteosynthesis and open reduction internal fixation on proximal humeral fracture in elder patients: a systematic review and meta-analysis. BioMed Research International 2017 2017 3431609. (https://doi.org/10.1155/2017/3431609)
- 46.↑
Xie L, Zhang Y, Chen C, Zheng W, Chen H, & Cai L. Deltoid-split approach versus deltopectoral approach for proximal humerus fractures: a systematic review and meta-analysis. Orthopaedics and Traumatology, Surgery and Research 2019 105 307–316. (https://doi.org/10.1016/j.otsr.2018.12.004)
- 47.↑
Haupt S, Weber S, Frima H, Hutter R, Grehn H, & Sommer C. Proximal humeral fracture-dislocation: outcome analysis in osteosynthesis and arthroplasties. European Journal of Orthopaedic Surgery and Traumatology: Orthopedie Traumatologie 2023 33 305–314. (https://doi.org/10.1007/s00590-021-03183-x)
- 48.↑
Robinson CM, Stirling PHC, Goudie EB, MacDonald DJ, & Strelzow JA. Complications and long-term outcomes of open reduction and plate fixation of proximal humeral fractures. Journal of Bone and Joint Surgery. American Volume 2019 101 2129–2139. (https://doi.org/10.2106/JBJS.19.00595)
- 49.↑
Haupt S, Frima H, & Sommer C. Operative treatment of proximal humeral fracture-dislocations through an anterolateral deltoid split approach. Archives of Bone and Joint Surgery 2020 8 589–597. (https://doi.org/10.22038/abjs.2020.42728.2162)
- 50.↑
Smith J, Berry G, Laflamme Y, Blain-Pare E, Reindl R, & Harvey E. Percutaneous insertion of a proximal humeral locking plate: an anatomic study. Injury 2007 38 206–211. (https://doi.org/10.1016/j.injury.2006.08.025)
- 51.↑
Traver JL, Guzman MA, Cannada LK, & Kaar SG. Is the axillary nerve at risk during a deltoid-splitting approach for proximal humerus fractures? Journal of Orthopaedic Trauma 2016 30 240–244. (https://doi.org/10.1097/BOT.0000000000000492)
- 52.↑
Ruchholtz S, Hauk C, Lewan U, Franz D, Kühne C, & Zettl R. Minimally invasive polyaxial locking plate fixation of proximal humeral fractures: a prospective study. Journal of Trauma 2011 71 1737–1744. (https://doi.org/10.1097/TA.0b013e31823f62e4)
