Deltopectoral vs deltoid split approach for proximal humerus fractures treated with locking plate: Our experience

Abstract

BACKGROUND

Proximal humerus fractures (PHFs) are common, especially in the elderly, and optimal surgical management remains debated. This study compares clinical, functional, and radiographic outcomes of deltoid split (DS) vs deltopectoral (DP) approaches in PHFs treated with locking plates.

AIM

To evaluate and compare the clinical, functional, and radiographic outcomes-as well as postoperative complication rates-associated with the DS vs the DP surgical approach in the open reduction and internal fixation (ORIF) of PHFs using locking plate constructs.

METHODS

A multicenter retrospective study of 120 patients undergoing ORIF for closed Neer type II-IV PHFs between January 2023 and December 2023. Patients were grouped by surgical approach [DS (n = 70), DP (n = 50)]. Outcome measures included Numeric Rating Scale (NRS) for pain, Quick-Disabilities in Arm, Shoulder, and Hand questionnaire (QuickDASH), Constant-Murley score, Short Form Health Survey-12v2, and radiographic alignment. Complication rates were recorded. Statistical significance was defined as P < 0.05.

RESULTS

Early outcomes favored the DS group: (1) Lower NRS (3.1 vs 5.9); (2) Higher Constant-Murley (68.2 vs 50.5); and (3) Better QuickDASH (25.4 vs 37.1). Complication rate was lower in the DS group (1.66% vs 5.81%). Radiographic outcomes were comparable. Long-term results were similar between groups.

CONCLUSION

While both approaches yield satisfactory long-term outcomes, the DS approach is associated with faster early recovery and fewer complications, supporting its use in selected cases.

Key Words: Proximal humerus fracture; Deltoid split; Deltopectoral; Locking plate; Open reduction and internal fixation; Functional outcome

Core Tip: This multicenter retrospective study compares deltoid split (DS) and deltopectoral approaches in proximal humerus fracture fixation with locking plates. While both methods achieved similar long-term results, the DS approach provided superior early functional outcomes and fewer complications. These findings suggest it may be the preferred technique in selected patients, particularly when early recovery is a priority.

INTRODUCTION

Proximal humerus fractures (PHFs) represent approximately 5% of all fractures in adults and are the third most common fracture type in individuals over 65 years old. These injuries show a bimodal distribution, with low-energy falls in elderly patients and high-energy trauma in younger individuals. Treatment options vary depending on the fracture pattern, bone quality, and patient comorbidities. Open reduction and internal fixation (ORIF) using locking plates has become a standard approach for displaced PHFs, providing stable fixation and allowing early mobilization[1]. Two primary surgical approaches are currently used in ORIF: (1) The deltopectoral (DP); and (2) Deltoid split (DS) approaches. The DP approach offers broad exposure of the proximal humerus and familiarity among surgeons, whereas the DS approach provides direct lateral access with minimal soft tissue dissection.

Despite the widespread use of both techniques, there is no clear consensus in the literature regarding the superiority of one approach over the other in terms of clinical, functional, and radiographic outcomes. Comparative studies and meta-analyses have yielded conflicting results, with some favoring the DS approach for early recovery, while others report similar long-term outcomes between groups. Furthermore, differences in complication patterns, particularly related to humeral head vascularity and nerve injury risk, are still debated. A key unresolved question is whether the choice of surgical approach significantly influences early recovery and complication rates in patients undergoing ORIF with locking plates.

Based on this gap in current knowledge, we conducted a retrospective multicenter cohort study to compare early and long-term clinical, radiological, and functional outcomes between the DP and DS approaches for displaced PHFs treated with locking plates. Our aim was to evaluate not only functional recovery and complication rates, but also to explore whether the choice of approach independently affects these outcomes.

MATERIALS AND METHODS

Study design and population

This was a retrospective multicenter cohort study conducted across three orthopedic departments in Italy between January 2023 and December 2023. A total of 120 patients with displaced PHFs were enrolled and treated surgically with ORIF using locking plate systems. All surgeries were performed by experienced shoulder surgeons with standardized protocols (Table 1).

Table 1 Baseline characteristics and surgical parameters, n (%).

Variable	Deltoid split	Deltopectoral	P value
Number of patients	70	50	-
Female	40 (57.1)	28 (56.0)	0.89
Male	30 (42.9)	22 (44.0)	0.89
Mean age (years)	56.0 ± 8.3	56.5 ± 7.9	0.76
Fracture type (Neer II/III/IV)	37/21/12	22/15/13	0.58
Side (right/Left)	38/32	27/23	0.97
Incision length (cm)	10.0 ± 1.2	13.0 ± 1.5	< 0.001
Operative time (minutes)	68.5 ± 10.3	70.2 ± 9.7	0.45
Follow-up duration (months)	14.6 ± 2.8	14.3 ± 2.7	0.65

Inclusion and exclusion criteria

Inclusion criteria were: (1) Age > 18 years; (2) Closed, displaced PHFs classified as Neer type II, III, or IV; (3) Treatment with ORIF using either a DS or DP approach; and (4) Minimum follow-up of 12 months.

Exclusion criteria were: (1) Open or pathological fractures; (2) Pre-existing shoulder arthropathy; (3) Previous surgery on the affected shoulder; (4) Polytrauma patients; and (5) Incomplete clinical or radiographic follow-up.

Surgical procedure

Patients were divided into two groups according to the surgical approach used: (1) 70 patients underwent the DS approach; and (2) 50 patients the DP approach. In both groups, anatomical locking plates were used, and fracture reduction was confirmed intraoperatively using fluoroscopy. The choice of surgical approach was made at the discretion of the operating surgeon based on preoperative fracture pattern, soft tissue condition, surgeon’s experience, and the expected ease of fracture reduction. In general, the DS approach was preferred for fractures involving the greater tuberosity or those requiring more lateral exposure, while the DP approach was selected in complex three-part or four-part fractures requiring more extensive medial visualization and calcar reconstruction. All procedures were performed in the beach chair position under general anesthesia.

Operative techniques

DP approach: The patient was positioned in the beach chair position with the arm draped freely. A skin incision was made along the DP groove, starting from the coracoid process and extending distally for approximately 12–15 cm. The DP interval was identified and developed with lateral retraction of the deltoid and medial retraction of the pectoralis major. The cephalic vein was preserved whenever possible. This approach provided excellent access to the anterior and medial humeral head, calcar, and lesser tuberosity, facilitating anatomical reduction in complex fractures. However, it required extensive soft tissue dissection and had limited exposure of the greater tuberosity.

DS Approach: In this technique, a skin incision of approximately 8–10 cm was made lateral to the acromion, centered over the deltoid muscle. The fibers of the anterior and middle deltoid were split bluntly in line with the muscle fibers, typically not extending more than 5 cm distal to the acromion to avoid injury to the axillary nerve. The axillary nerve was identified and protected when necessary. This approach allowed direct access to the lateral aspect of the proximal humerus, making it particularly useful for fractures involving the greater tuberosity. The DS approach minimized soft tissue disruption and preserved the anterior vascular supply but offered more limited visualization of the medial calcar.

Clinical and radiographic evaluation

Clinical outcomes were assessed using the following validated scoring systems.

Constant-Murley Score: A composite score evaluating shoulder function across four domains: (1) Pain (15 points); (2) Daily activities (20 points); (3) Range of motion (40 points); and (4) Strength (25 points). The maximum score is 100, with higher scores indicating better function[2].

Quick-Disabilities in Arm, Shoulder, and Hand questionnaire: An 11-item self-reported questionnaire measuring upper limb disability and symptoms. Scores range from 0 (no disability) to 100 (severe disability)[3].

Numeric Rating Scale for pain: Patients rate their pain on a scale from 0 (no pain) to 10 (worst imaginable pain), providing a simple and reliable assessment of pain intensity[4].

Short Form Health Survey-12v2: A standardized tool to assess health-related quality of life across physical and mental domains. It generates two summary scores (Physical Component Summary and Mental Component Summary), with higher values representing better perceived health status.

Radiographic analysis included: (1) Measurement of head-shaft angle; (2) Assessment of medial calcar restoration; and (3) Detection of complications such as implant failure, avascular necrosis (AVN), and nonunion.

Postoperative care and rehabilitation

All patients followed a standardized postoperative protocol supervised by a physiotherapist. The operated arm was immobilized in a sling for comfort for the first 10–14 days. Passive range-of-motion exercises, including pendulum movements, were initiated on postoperative day one, unless contraindicated by fracture complexity or poor fixation stability.

From week 3 to 6, assisted active motion exercises were introduced, focusing on forward flexion and external rotation. Strengthening exercises targeting the rotator cuff and deltoid muscles began after radiographic evidence of healing, typically after 6–8 weeks.

Patients operated via the DP approach often required a slightly more cautious progression due to more extensive soft tissue dissection, particularly during early abduction movements. In contrast, those undergoing the DS approach generally tolerated earlier lateral movement but required careful monitoring to avoid axillary nerve irritation.

Clinical and radiographic follow-ups were scheduled at 2 weeks, 6 weeks, 3 months, 6 months, and 12 months postoperatively.

Statistical analysis

Continuous variables were reported as mean ± SD, while categorical variables were expressed as frequencies and percentages. The Shapiro–Wilk test was used to assess normality. Between-group comparisons were performed using the Students’ t-test or Mann–Whitney U test for continuous data and the χ² or Fisher’s exact test for categorical variables. A P < 0.05 was considered statistically significant. Data were analyzed using Statistical Package for the Social Sciences v26.0 (IBM Corp., Armonk, NY, United States) (Table 2).

Table 2 Summary of study design, groups, and variables.

Parameter	Description
Study design	Multicenter retrospective cohort study
Study period	January 2023–December 2023
Total patients	120
Surgical approaches	Deltoid split (n = 70), deltopectoral (n = 50)
Inclusion criteria	Age > 18, Neer type II–IV, closed fracture, open reduction and internal fixation, 12-month follow-up
Exclusion criteria	Open/pathological fractures, prior surgery, incomplete data, polytrauma
Implant type	Anatomical locking plate
Outcomes measured	Constant-Murley, Quick-Disabilities in Arm, Shoulder, and Hand questionnaire, Numeric Rating Scale, Short Form Health Survey-12v2
Radiographic evaluation	Head-shaft angle, calcar restoration, complications
Statistical analysis	t-test, Mann–Whitney U, χ2, Fisher’s exact; P < 0.05

RESULTS

A total of 120 patients were included in the study, with 70 (58.3%) treated using the DS approach and 50 (41.7%) treated via the DP approach. Baseline demographics and fracture characteristics were similar between groups (Table 3). According to the Neer classification, the majority of patients presented with type II and III fractures in both cohorts (Table 4).

Table 3 Patient distribution by surgical approach.

Group	Number of patients	Percentage
Deltoid split	70	58.3
Deltopectoral	50	41.7

Table 4 Fracture types according to Neer classification.

Group	Neer II	Neer III	Neer IV
Deltoid split	37	21	12
Deltopectoral	22	15	13

Surgical parameters

The mean incision length was significantly shorter in the DS group (10 ± 1.2 cm) compared to the DP group (13 ± 1.5 cm). Mean operative time was 68.5 ± 10.3 minutes for DS and 70.2 ± 9.7 minutes for DP (no statistical significance of P value). All procedures were performed under general anesthesia, with the patient in the beach chair position. No intraoperative complications were reported. Postoperative radiographs confirmed an adequate reduction in all cases, defined as head-shaft angle 130° ± 10° with calcar restoration.

Early functional outcomes

At 1-month follow-up, pain levels and functional scores favored the DS group. The DS group had a lower mean Numeric Rating Scale (NRS) score (6.0 ± 1.4) compared to the DP group (9.0 ± 1.5, P < 0.001), and a higher Constant-Murley score (39 ± 6.8 vs 30 ± 7.1, P = 0.004). Quick-Disabilities in Arm, Shoulder, and Hand questionnaire (QuickDASH) scores were also slightly better in the DS group (45.2 ± 8.7 vs 48.3 ± 7.5, P = 0.09), although not statistically significant. At 3 months, intergroup differences became more pronounced. The DS group showed: (1) Lower NRS scores (3.1 ± 1.2 vs 5.9 ± 1.3, P < 0.001); (2) Higher Constant-Murley scores (68.2 ± 8.1 vs 50.5 ± 9.4, P = 0.002); (3) Lower QuickDASH scores (25.4 ± 7.8 vs 37.1 ± 6.9, P < 0.001); and (4) Lower Short Form Health Survey (SF)-12v2 scores (32.6 ± 6.3 vs 40.3 ± 5.4, P = 0.003) (Table 5, Figure 1).

Figure 1 Numeric Rating Scale pain score, Quick-Disabilities in Arm, Shoulder, and Hand questionnaire Score, and Constant-Murley score over time. A: Numeric Rating Scale Pain score; B: Quick-Disabilities in Arm, Shoulder, and Hand questionnaire score; C: Constant-Murley score. NRS: Numeric Rating Scale; QuickDASH: Quick-Disabilities in Arm, Shoulder, and Hand questionnaire.

Table 5 Statistical comparison at 3-month follow-up.

Parameter	Deltoid split (mean ± SD)	Deltopectoral (mean ± SD)	P value
Numeric Rating Scale Score	3.1 ± 1.2	5.9 ± 1.3	< 0.001
Quick-Disabilities in Arm, Shoulder, and Hand questionnaire	25.4 ± 7.8	37.1 ± 6.9	< 0.001
Short Form Health Survey-12v2	32.6 ± 6.3	40.3 ± 5.4	0.003
Constant-Murley	68.2 ± 8.1	50.5 ± 9.4	0.002

Long-term outcomes

By the 6-month follow-up and beyond, differences between the groups diminished. Both groups demonstrated favorable clinical outcomes: (1) Constant-Murley: DS: 93 ± 4.2, DP: 91 ± 5.1 (P = 0.32); (2) QuickDASH: DS: 13.2 ± 4.6, DP: 14.1 ± 5.2 (P = 0.47); and (3) NRS: 1.0 ± 0.7 in both groups (P = 0.90). These findings indicate that while early recovery is accelerated in the DS group, both approaches ultimately converge to similar long-term results (Table 6).

Table 6 Clinical outcomes at follow-up.

Time point	Numeric Rating Scale (DS/DP)	Quick-Disabilities in Arm, Shoulder, and Hand questionnaire (DS/DP)	Short Form Health Survey-12v2 (DS/DP)	Constant-Murley (DS/DP)
1 month	6/9	45/48	47/52	39/30
3 months	3/6	25/37	32/40	68/50
6+ months	1/1	13/14	16/15	93/91

DP: Deltopectoral; DS: Deltoid split.

Postoperative complications

The overall complication rate was significantly lower in the DS group (1.66%) compared to the DP group (5.81%, P = 0.04) (Table 7, Figure 2). In the DS group, one patient developed a superficial wound infection and one experienced secondary loss of reduction, though clinical function was preserved. In the DP group, complications included: (1) Two superficial infections; (2) One deep infection requiring hardware removal and local antibiotic application; (3) Two cases of AVN of the humeral head; and (4) Two screw cut-outs requiring surgical revision. No cases of neurovascular injury were recorded in either group. A total of 120 patients met the inclusion criteria and were enrolled. Of these, 70 (58.3%) were treated using the DS approach, and 50 (41.7%) with the DP approach. Groups were homogeneous in terms of sex, age, and fracture classification. In the DS group, there were 40 females and 30 males (mean age 56), with 37 Neer II, 21 Neer III, and 12 Neer IV fractures. The DP group included 28 females and 22 males (mean age 56.5), with 22 Neer II, 15 Neer III, and 13 Neer IV fractures. Surgery duration was similar, though incisions were shorter in the DS group (10 cm vs 13 cm in DP). Radiographic evaluation confirmed adequate reduction in all cases. Pain and functional scores were better in the DS group during early follow-up. At 1 month, NRS was 6 in DS and 9 in DP; Constant-Murley was 39 vs 30; QuickDASH was 45 vs 48. At 3 months, DS patients improved more rapidly: (1) NRS was 3 vs 6; (2) Constant-Murley was 68 vs 50; (3) QuickDASH was 25 vs 37; and (4) SF-12v2 was 32 vs 40. At later follow-ups, outcomes were comparable (Constant-Murley was 93 vs 91, QuickDASH was 13 vs 14, NRS was 1 in both). Complications were lower in the DS group (1.66%) than in DP (5.81%). DS had one superficial infection and one loss of reduction without clinical impact. DP had two superficial infections, one deep infection requiring implant removal, two AVN cases, and two screw cut-outs needing revision surgery.

Figure 2 Postoperative complications by group. AVN: Avascular necrosis.

Table 7 Postoperative complications.

Complication	Deltoid split (n)	Deltopectoral (n)
Superficial wound infection	1	2
Deep infection	0	1
Loss of reduction	1	0
Avascular necrosis	0	2
Screw cut-out	0	2

Subgroup analysis

We performed a subgroup analysis to evaluate whether clinical and functional outcomes differed based on patient age, sex, fracture type (Neer classification), side of injury, and incision length. No statistically significant interaction was observed between these variables and the choice of surgical approach.

Age and sex: Both younger and older patients achieved similar improvements in functional scores within each group, with no significant interaction between age or sex and surgical outcomes (P > 0.05).

Fracture type: Patients with Neer II, III, and IV fractures showed consistent outcome trends within each surgical group, with the DS approach maintaining superior early functional scores irrespective of fracture complexity.

Side of injury: No difference in outcomes was observed between right-sided and left-sided fractures within or between groups.

Incision length: While incision length differed significantly between approaches, it did not correlate with postoperative pain or functional recovery. Data on plate length were not systematically recorded and thus could not be included in the comparative analysis.

DISCUSSION

Numerous studies have compared the DP and DS approaches in the surgical treatment of PHFs, yet no definitive consensus has been reached regarding their relative efficacy. Several systematic reviews and meta-analyses, including those by Wu and Shen[5] and Xie et al[6], have demonstrated that the DS approach may offer advantages in terms of shorter operative time, reduced intraoperative blood loss, and improved early functional outcomes-particularly higher Constant-Murley scores at 3 months. These studies also noted that long-term outcomes such as pain scores and rates of AVN were generally comparable between approaches.

Conversely, randomized controlled trials have shown conflicting results. Rouleau et al[7] reported better long-term QuickDASH and SF-12v2 scores with the DP approach, while Basavanagowda et al[8] found no significant differences between the two techniques in long-term follow-up. Such inconsistencies highlight the complex interplay of patient factors, fracture type, and surgeon preference in determining optimal outcomes. PHFs are increasingly common due to aging populations. Palvanen et al[9] showed an increase of incidence from 32/100000 to 105/100000 from 1970 to 2002. The better treatment is still debate in literature. Iyengar et al[10] showed that nonoperative treatment of PHFs demonstrates high rates of radiographic healing, good functional outcomes, and a modest complication rate. Syed et al[11] in their systematic review and meta-analysis recommend nonoperative treatment for the average elderly patient (aged > 65 years) with a displaced proximal humeral fracture. Surgical treatment is often warranted for displaced or unstable fractures. Hohmann et al[12] declared that ORIF is superior to hemiarthroplasty, ORIF is comparable to intramedullary nailing and to reverse shoulder arthroplasty. Despite advances in fixation techniques, the optimal surgical approach remains under debate. Yahuaca et al[13] found no difference between older patients with high American Society of Anesthesiologists (ASA) treated with arthroplasty, and younger patients with lower ASA were treated with ORIF; both groups showed improvement in function at 1 year of follow up. Our study compared two widely used approaches-DS and DP-in patients treated with locking plates. While both approaches produced satisfactory long-term results, the DS group showed significantly faster early recovery and a lower complication rate. These advantages may stem from the DS approach's more direct lateral access, reduced soft tissue disruption, and better preservation of vascular supply. All AVN cases occurred in the DP group, consistent with literature suggesting potential compromise of the humeral head’s vascularity with this approach. However, Bhayana et al[14] noted similar results for AVN, malunion and clinical scores in both group at last follow up. Our findings align with Wu and Shen[5] and Hepp et al[15], who reported similar trends. Rouleau et al[7], however, found better functional scores in DP, albeit with higher complication rates. Buecking et al[16] observed no significant differences but reported different complication patterns between the two approaches. Kishore et al[17] didn’t found differences in both DP and DS group in terms of blood loss, surgical time, high anatomical reduction and complications. Liu et al[18] in their paper noted a better functional score at 3 months and 6 months follow up in DS group than DS group, but they didn’t found differences in both groups at 12 months follow up. Schliemann et al[19] described varus malunion as the most common complication after locking plate of proximal humeral fracture; others complications are: (1) AVN; (2) Screw cut-out; and (3) Subacromial impingement and infection. Ruchholtz et al[20] declared that the different rate of complications in literature depends on surgeon’s experience. Wang et al[21] compared outcome between mini-invasive DS and extensive DS and they noted less intraoperative X-ray and surgical time in extensive group than mini-invasive group vas score and hospital stay were longer in extensive DS group. Mipo group showed a higher secondary reduction loss than extensive DS group. Moreover, Handoll et al[22] noted that a problem frequently discussed and debated in literature during placement of plate and screws during the mini open DS approach is damage of circumflex nerve with crippling sequele such as abduction deficit and a strong limitation of muscular strength. Sirisreetreerux et al[23] in their cadaveric study compared exposure area of different surgical approach; they found a major exposure area in the inverted L anterolateral deltoid flip but they not found differences between DS an DP surgical approach. Importantly, we minimized nerve injury risk in DS by using longer incisions and direct visualization of the axillary nerve. We also reinforced stability with rotator cuff sutures and proper calcar screw placement, techniques supported by Kimmeyer et al[24]. In conclusion, while both approaches are valid, DS may offer early recovery benefits and fewer complications in experienced hands.

Strengths and limitations

Strengths: This study benefits from a relatively large and homogenous sample size for a surgical cohort, with patients treated in multiple high-volume orthopedic centers following standardized surgical and postoperative protocols. Additionally, the use of validated outcome measures and long-term clinical and radiographic follow-up (up to 36 months) enhances the reliability of the findings.

Limitations: As a retrospective study, it is inherently subject to selection and information bias. Despite efforts to standardize treatment, surgical decision-making and technique may have varied slightly between centers. Moreover, the lack of randomization limits the ability to establish causality. The absence of preoperative functional scores and patient-reported outcome measures also restricts the baseline comparability. Finally, although complication rates were recorded, specific time points and severity gradings were not uniformly documented.

CONCLUSION

This multicenter retrospective cohort study demonstrates that both the DP and DS approaches for proximal humerus fracture fixation with locking plates can achieve comparable long-term functional outcomes. However, the DS approach was associated with significantly faster early recovery, lower pain scores, and a reduced complication rate-particularly concerning AVN and implant-related failures. These findings suggest that the choice of surgical approach does influence early postoperative trajectories, likely due to differences in soft tissue handling, vascular preservation, and ease of lateral exposure for implant positioning. The DS approach, when performed with meticulous technique and adequate nerve protection, represents a valuable option in selected patients, particularly in those where early mobilization and functional recovery are priorities. Future prospective, randomized controlled trials with larger cohorts are warranted to confirm these results and to further elucidate the specific patient populations that might benefit most from each surgical approach.