Intramedullary nailing versus plate fixation for humeral shaft fractures in geriatric patients: a retrospective cohort study assessing functional outcomes and complication rates

Abstract

Background

Humeral shaft fractures, accounting for 1–3% of adult fractures, increasingly require surgical intervention due to high nonunion rates. This study compares outcomes of intramedullary nailing and plate fixation in older adults.

Methods

A retrospective analysis was conducted on 55 patients with humeral shaft fractures treated at the Mohamed Taieb Kassab Orthopedic Institute between 2014 and 2022. The cohort included 32 patients managed with intramedullary nailing (IMN) and 23 with dynamic compression plating (DCP). Demographics, fracture characteristics (AO classification), operative duration, union duration, functional outcomes (modified Stewart and Hundley criteria, Constant-Murley score), and complications were evaluated. Statistical analysis used SPSS v19.0.

Results

Mean age was 64.7 years, with male predominance (40 male, 15 female and a gender ratio of 2.6). IMN was preferred for middle-third (58%) and AO type A fractures (79.4%), while DCP dominated complex fractures (76.2% type B/C with p < 0.05).IMN demonstrated shorter operative duration (60.2 ± 4.4 vs. 105.9 ± 7.7 min and p < 0.001). Fracture union averaged 19.9 ± 7.9 weeks, with no intergroup difference (p = 0.360). Functional outcomes (76.4% good or excellent) were comparable between techniques (p > 0.05), but DCP yielded superior Constant-Murley scores (91.1 ± 10.5 vs. 80.6 ± 15.4 and p = 0.007).

Complications included shoulder stiffness (25% IMN, 0% DCP and p = 0.015), complex regional pain syndrome (CRPS) (15.6% IMN, 0% DCP), sepsis (4.3% DCP, 0% IMN), and radial nerve palsy (4.3% DCP, 0% IMN).

Conclusion

IMN offers faster surgery and lower infection risk but higher shoulder morbidity. Plate fixation excels in complex fractures with better functional scores but longer operative duration and infection risk. Considering these factors, DCP may be the preferred approach in older adults compared to IMN, particularly when functional optimization is a priority. The choice of surgical technique should be guided by fracture characteristics, patient’s specific factors, and surgeon expertise.

Background

Humeral shaft fractures, which occur between the insertion of the pectoralis major muscle and the distal insertion of the brachialis muscle, account for 1–3% of all fractures in adults [1, 2]. These fractures can significantly impact the functional prognosis of the upper limb, particularly in older adults, requiring prompt and appropriate management [3, 4]. The incidence of these fractures has been rising, largely due to an increase in road traffic accidents [5, 6].

While the diagnosis is straightforward, relying on clinical and radiological examinations, the choice of treatment remains controversial and often depends on the surgeon's preference and experience [7].

Traditionally, non-operative treatment was considered the gold standard due to the non-weight-bearing nature of the humerus and satisfactory outcomes [8]. However, the high rate of nonunion, up to 18% in recent series, has led to a shift towards surgical intervention [8]. The two primary surgical techniques are intramedullary nailing (IMN) and plate fixation (PF), each with its own set of advantages and complications.

Currently, there are few studies comparing the efficacy of these two techniques in the treatment of humeral shaft fractures in older population. To address this knowledge gap, we conducted a retrospective study to compare the clinical and radiological outcomes of IMN and PF using dynamic compression plating (DCP) in the treatment of humeral shaft fractures.

Materials and methods

We retrospectively analyzed data collected from patients with humeral shaft fractures treated surgically at the Mohamed Taieb Kassab Orthopedic Institute between January 2014 and December 2022.

Inclusion and exclusion criteria

Inclusion criteria included patients older than 60 years of age with humeral shaft fractures and patients treated whether with PF or IMN, while exclusion criteria involved pathological fractures, metaphyseal-epiphyseal extension, distal third fractures, preoperative shoulder dysfunction, open fractures, initial radial nerve palsies and incomplete records.

Patient selection

A retrospective review of 96 medical records was conducted. After applying the inclusion and exclusion criteria, 55 patients were included in the study. These patients were divided into two groups based on the osteosynthesis technique:

• IMN group: Antegrade second-generation IMN fixation.

• DCP group: Plate fixation using DCP.

Surgical techniques

For IMN group, an anterolateral approach centered on the acromion tip was used, splitting the deltoid muscle along its fibers. The entry point was located at the posterior aspect of the greater tuberosity, posterior to the long head of the biceps tendon and at the osteochondral junction. A curved proximal-end nail with bipolar locking was employed for fixation.

For DCP group, plate fixation was performed through an anterolateral approach.

Surgical technique selection in this retrospective study was determined by surgeon preference alone, as no standardized criteria or documentation of decision rationale were available.

Data collection

The collected data encompassed demographic information, fracture characteristics classified using the Arbeitsgemeinschaft für Osteosynthesefragen (AO) system [9], surgical details (including operative duration and approach), and postoperative outcomes (such as union rates and complications).

Follow-up protocol

Patients were assessed clinically and radiographically at 6 weeks, 3 months, 6 months, and 12 months postoperatively, with additional visits until union was confirmed.

Outcome measures

The assessment of radiographic union, functional outcomes and complications was conducted independently by multiple orthopedic surgeons.

Primary outcome

Fracture union:

Fracture Union was defined as the presence of bridging callus across the fracture site within 6 months postoperatively. Nonunion was characterized by persistent fracture lines visible beyond 9 months with no evidence of progressive healing over 3 consecutive months of follow-up [10].

Secondary outcomes

Functional outcomes and complications:

Functional outcomes were assessed using the modified Stewart and Hundley classification [10, 11] and the Constant- Murley score.

The Constant–Murley score was utilized to assess shoulder joint function. This scoring system has a maximum of 100 points, distributed as follows: 15 points for pain, 20 points for daily living activities, 40 points for range of motion, and 25 points for muscle strength. Shoulder function is classified based on the total score: 90–100 points indicate excellent function, 80–89 points signify good function, 70–79 points reflect moderate function, and a score below 70 points denotes poor function [12].

Shoulder stiffness was defined as restricted range of motion with < 100° of forward flexion or < 30° of external rotation [13].

Postoperative residual pain was assessed using the established criteria embedded within the modified Stewart and Hundley classification system. This tool provided a structured framework for categorizing pain based on patient-reported symptoms and clinical evaluation. Specifically, pain was classified into one of four distinct categories: “None,” indicating a complete absence of pain, “Occasional,” referring to minimal pain that arises infrequently, such as during weather changes or strenuous activity, “After effort,” describing pain elicited by specific physical activities, and “Permanent,” designating persistent pain that is present even at rest or during daily activities.

The assessment of outcomes, including the application of the Stewart and Hundley classification and the Constant-Murley score, was conducted by the attending orthopedic surgeon present at each scheduled follow-up appointment within our department.

To demonstrate the technical approach and results, representative examples of each surgical technique are presented in Figs. 1 and 2.

Fig. 1.

An example case of a right midshaft humeral fracture classified as AO type A3 in a 62-year-old female patient treated with plate fixation using a dynamic compression plate (DCP). A X-ray image showing a right midshaft humeral fracture; (B) Postoperative anteroposterior X-ray image showing satisfactory reduction; (C) Anteroposterior X-ray at 8 months postoperatively showing good healing with no displacement of the fracture

Fig. 2.

An example case of a right midshaft humeral fracture classified as AO type B1 in a 60-year-old male patient treated with intramedullary nailing. A Preoperative anteroposterior X-ray showing a right midshaft humeral fracture. B Postoperative anteroposterior X-ray showing satisfactory reduction and intramedullary nail fixation. C Anteroposterior X-ray at 4 years postoperatively showing complete bone healing with no displacement of the fracture

Statistical analyses were conducted using SPSS (version 19.0). Normality was assessed using the Shapiro–Wilk test. For normally distributed data, Student’s t-test was applied. For non-normally distributed data, the Mann–Whitney U test was used. Categorical variables were evaluated with the chi-square test. Results are reported as mean ± standard deviation for normally distributed variables, and median (interquartile range) for non-normal variables. A p-value of < 0.05 was considered statistically significant.

Results

A total of 55 cases were divided into two groups based on the osteosynthesis technique used: 32 patients were treated with antegrade second generation IMN with a curved proximal end and bipolar locking, and 23 patients were treated with DCP. The mean age of the patients was 64.7 ± 4.9 years. The demographic distribution was significantly skewed towards males (72.7%, n = 40), with females representing 27.3% (n = 15) of the 55 participants, equivalent to a gender ratio of 2.6.

Of the study population, 67.3% (37 patients) were admitted following road traffic accidents. Isolated humeral shaft fractures were seen in 47 patients, while 18 cases (32.7%) involved lower-energy trauma.

The dominant side was affected in 58.2% of cases, with no statistically significant difference (p = 0.16). The fracture mechanism was direct in 50.9% of cases (28 patients).

The middle third of the humeral shaft was the most common fracture site, accounting for 90.9% of cases. According to the AO classification, type A fractures were the most frequent, occurring in 34 cases (61.2%), followed by type B in 18 cases (32.7%) and type C in 3 cases (5.4%).

The mean follow-up duration for the study cohort was 12.7 months, with a range from 10 to 17 months. The IMN group had a mean follow-up of 13 months, compared to 12.3 months for the DCP group (p = 0.153).

Baseline patient characteristics are summarized in Table 1.

Table 1.

Comparison of baseline patient characteristics

Characteristic	IMN (n = 32)	DCP (n = 23)	P-value
Age (years)	63.6 ± 4.2	66.6 ± 5.6	0.042
Sex, n (%)			0.363
Male	25 (78.1%)	15 (65.2%)
Female	7 (21.9%)	8 (34.8%)
Fracture type (AO), n (%)			< 0.05
A	27 (84.4%)	7 (30.4%)
B	5 (15.6%)	13 (56.5%)
C	0	3 (13%)
Fracture site, n (%)			1
Upper third	3 (9.4%)	2 (8.7%)
Middle third	29 (90.6%)	21 (91.3%)

Surgical outcomes

Operative duration was significantly shorter for IMN group (60.2 ± 4.4 min) compared to DCP group (105.9 ± 7.7 min and p < 0.001). All type C fractures were treated with DCP. Of the fractures in the middle third, 58% (29 cases) were treated with IMN, compared to 21 cases treated with DCP. Type A fractures were treated with IMN in 79.4% of cases, while DCP was used in 76.2% of type B or C fractures, with a statistically significant difference (p < 0.05).

Fracture union and functional outcomes

Mean time to union was 19.9 ± 7.9 weeks, with 19.1 ± 8.6 weeks in the IMN group and 21.1 ± 6.8 weeks in the DCP group (p = 0.360).

Postoperative residual pain affected 10 patients (18.2%), distributed between surgical groups: nine in the IMN group and 1 in the DCP group with a statistically significant difference favoring DCP (p = 0.034).

Functional outcomes assessed via the modified Stewart and Hundley classification demonstrated good or excellent results in 76.4% of patients overall, irrespective of technique (p > 0.05).

While 80% of upper-third fractures achieved good or excellent functional outcomes, middle-third fractures had a slightly lower rate (76%), with p > 0.05.

Simple type A fractures demonstrated a higher rate of good-to-excellent outcomes compared to other fracture types, though this difference did not reach statistical significance (p = 0.059).

Functional assessment (Constant-Murley Scale)

Functional outcomes assessed using the Constant-Murley scale were significantly better in DCP group (mean score: 91.1 ± 10.5 vs. 80.6 ± 15.4 and p = 0.007).

These findings, along with detailed postoperative functional outcomes, are summarized in Table 2.

Table 2.

Fracture union and functional outcomes

Outcomes	Variable	Total (n = 55)	IMN group (n = 32)	DCP group (n = 23)	p-value
	Mean time to union (weeks)	19.9 ± 7.9	19.1 ± 8.6	21.1 ± 6.8	0.360
	Postoperative residual pain	18.2% (10)	28.1% (9)	4.3% (1)	0.027
	Good or excellent results	76.4% (42)	75% (24)	78.3% (18)	1
	Mean Constant-Murley scale	85 ± 14.5	80.6 ± 15.4	91.1 ± 10.5	0.007

Complications

Postoperative complications varied between groups, as summarized in Table 3.

Table 3.

Complications of the IMN and DCP groups

Complication	Total (n = 55)	IMN group (n = 32)	DCP group (n = 23)	p-value
Shoulder stiffness	14.5% (8)	25% (8)	0% (0)	0.015
CRPS	9.1% (5)	15.6% (5)	0% (0)	0.068
Sepsis	1.8% (1)	0% (0)	4.3% (1)	0.418
nonunion	1.8% (1)	3.1% (1)	0% (0)	1
iatrogenic radial nerve palsy	1.8% (1)	0% (0)	4.3% (1)	0.418

In the IMN group, the most common complication was shoulder stiffness, affecting 25% (8/32) of patients, with a statistically significant difference compared to DCP group (0/23) (p = 0.015). Additionally, complex regional pain syndrome (CRPS) was observed in 8.1% (5/32) in IMN group, whereas no cases occurred in the DCP group (p = 0.068). One case of nonunion (3.1%) in the IMN group required revision surgery with DCP.

In the DCP group, complications included sepsis (4.3%, 1/23) and iatrogenic radial nerve palsy in one case (4.3%, 1/23) with a partial recovery at 3 months. However, no significant differences were found between the groups for these complications (p > 0.05).

Discussion

While the surgical management of humeral shaft fractures is widely debated, a significant gap persists in the literature regarding the optimal treatment for the vulnerable geriatric population. This gap is particularly pronounced in the Middle East and North Africa (MENA) region, where demographic trends and injury patterns may yield unique outcomes.

Our study is the first of its kind within the MENA region to specifically address this critical issue in elderly patients. This retrospective analysis of 55 geriatric patients provides crucial, region-specific data by comparing the outcomes of Intramedullary Nailing (IMN) and Dynamic Compression Plating (DCP).

The study cohort exhibited a distinct demographic and clinical profile, with a mean age of 64.7 years (± 4.9). A significant male predominance was observed, as males comprised 40 of the 55 patients (72.7%), resulting in a gender ratio of 2.6.

Fracture localization analysis revealed that the middle third of the humeral shaft was involved in the vast majority of cases (90.9%, n = 50).

According to the AO classification system, simple type A fractures were the most frequent (61.2%, n = 34), followed by type B (32.7%, n = 18) and type C (5.4%, n = 3) fractures.

Treatment selection was influenced by both fracture morphology and surgeon preference. This was evidenced by the preferential use of IMN for simple type A fractures (79.4%), while DCP was primarily reserved for comminuted patterns, accounting for 76.2% of type B and C fractures.

Operative efficiency substantially favored the IMN group, which demonstrated a 45-min reduction in mean surgical duration compared to DCP (60.2 ± 4.4 vs. 105.9 ± 7.7 min; p < 0.001). Radiographic healing timelines, however, were comparable between groups (19.9 ± 7.9 weeks; p = 0.360).

Functional outcomes

Functional recovery is a critical measure of success in the treatment of humeral shaft fractures. In our study, 76.4% of patients achieved good or excellent results according to the modified Stewart and Hundley classification, with no significant difference between IMN and PF groups. This is consistent with findings from other studies, which report comparable functional outcomes for both techniques [14, 15].

However, some studies suggest that PF may offer superior functional outcomes, particularly in terms of shoulder function. For example, Rabari et al. found that patients treated with PF had significantly better Disabilities of the Arm, Shoulder, and Hand (DASH) scores compared to those treated with IMN [16]. Similarly, Akalın et al. reported better Constant scores for PF, although the difference was not statistically significant [17].

In a 2022 meta-analysis comparing IMN and PF for humeral shaft fractures, Hu et al. reported that IMN was significantly inferior to PF in terms of the American Shoulder and Elbow Surgeons (ASES) scores (p = 0.04) and rates of shoulder or elbow functional limitations (p = 0.03), suggesting less favorable early postoperative functional recovery with IMN [18].

Current literature remains inconclusive regarding the optimal surgical technique for superior clinical outcomes for elderly population. However, recent meta-analyses lean toward PF over IMN due to the latter’s association with rotator cuff morbidity, particularly when using curved proximal-end nails that disrupt tendon integrity [19, 20].

To address these limitations, some authors propose using straight, small-diameter nails equipped with dynamic compression systems, which traverse the rotator cuff muscles without damaging tendons. Recent clinical studies, including work by Lopiz et al., demonstrate improved functional recovery with these modified nails compared to conventional designs [13].

Complications

Infection

Infection rates were low in our series, with one case of sepsis in the DCP group and none in the IMN group. This is consistent with the literature, which generally reports lower infection rates for IMN due to its closed reduction technique [21, 22]. However, multiple studies have reported that PF for humeral shaft fractures is associated with a higher risk of surgical site infection compared to IMN, particularly for fractures located in the middle and lower thirds [2, 14, 15, 19, 23].The observed difference in infection rates between PF and IMN groups did not reach statistical significance.

This difference is attributed to the extensive surgical exposure, greater intraoperative blood loss, and disruption of the fracture hematoma during plate application, which may predispose to bacterial colonization.

To mitigate infection risks in PF, some authors advocate for minimally invasive plate osteosynthesis (MIPO), which preserves periosteal blood supply and reduces soft-tissue dissection [23].

Nonunion and delayed union

Nonunion and delayed union are significant concerns in the treatment of humeral shaft fractures. In our study, one case of nonunion occurred in the IMN group, while no cases were reported in the PF group. These findings are consistent with other reports of increased nonunion risk following IMN, particularly among elderly individuals with compromised bone density or inadequate reduction [16, 24].

Plate fixation, with its ability to achieve anatomical reduction and rigid fixation, is generally associated with lower rates of nonunion [19, 23].

In patients with osteoporosis, the risk of delayed union is higher, particularly with IMN, due to compromised bone quality. Furthermore, fractures located in the proximal diaphysis carry a higher risk of delayed union, regardless of the fixation method employed. This is likely attributable to reduced vascular supply in this anatomical region [25, 26].

Shoulder stiffness

Shoulder stiffness emerged as the most frequent complication in the IMN group, with a prevalence of 25% (8/32). This incidence was significantly higher than in the DCP group, in which no cases were observed (0/23; p = 0.015).

This disparity is pathophysiologically linked to the violation of the rotator cuff during the antegrade nailing approach. The transgression of the supraspinatus tendon and its surrounding structures during nail insertion can incite a localized inflammatory response, leading to peritendinous fibrosis, adhesion formation within the subacromial space, and subsequent scarring [14]. This process directly compromises the biomechanics of the glenohumeral joint, often resulting in pain, weakness, and a characteristic restriction of motion, particularly in abduction and internal rotation [24]. The resulting shoulder stiffness is a significant contributor to functional impairment and prolonged rehabilitation following IMN [22].

In stark contrast, PF with a plate via an anterolateral approach meticulously spares the rotator cuff musculature. The dissection occurs between the deltoid and pectoralis major muscles, providing direct access to the humeral shaft without disturbing the critical tendinous structures of the rotator cuff [23].

The clinical impact of this anatomical distinction is profound. The absence of rotator cuff injury in the DCP group is a key factor explaining their superior functional outcomes, as quantitatively reflected in the significantly higher Constant-Murley scores documented in our study. This finding is consistent with a growing body of literature that links rotator cuff integrity to optimal shoulder function after humeral fixation [18, 23, 25]. Consequently, when selecting a surgical technique for humeral shaft fractures, particularly in active patients where shoulder function is paramount, the risk of iatrogenic stiffness from rotator cuff violation must be carefully weighed against the potential benefits of a minimally invasive IMN procedure.

Iatrogenic radial nerve palsy

One case of iatrogenic radial nerve palsy occurred in the DCP group in a 63-year-old female with a middle-third B3-type humeral fracture following a road traffic accident.

This case aligns with the well-documented higher risk of nerve injury in complex, comminuted fractures (AO Types B and C) of the middle third, where the radial nerve's anatomy makes it particularly vulnerable to entrapment within fracture fragments, stretching, or tethering by scar tissue [27, 28]. Based on this finding and existing literature, we recommend that surgeons exercise particular caution with comminuted fractures (AO Type B and C) in the middle third of the humerus, especially those resulting from high-energy trauma. To prevent iatrogenic palsy during plate fixation, we emphasize:

• Meticulous preoperative planning: Review imaging to identify comminution and anticipate the nerve's potential path.

• Adequate exposure and visualization: Employ an extensile approach to formally identify and protect the radial nerve proximal and distal to the fracture zone before manipulating fragments.

• Minimal and careful dissection: Avoid aggressive retraction. When working anterolaterally, meticulous dissection of the brachialis muscle is crucial to avoid injuring the radial nerve, which innervates part of it.

The radial nerve’s course relative to plate holes should be documented in the operative report to minimize injury during future plate removal [29].

Limitations

This study has several limitations. First, its monocentric and retrospective cohort design limits the generalizability of the findings. The small sample size limits the ability to detect sublet different between groups along with non-randomized distribution of surgical techniques that may introduce a significant potential for selection bias. Furthermore, the variability in surgical technique among the multiple participating surgeons, while reflective of real-world practice, introduces an additional source of bias that could influence outcomes and results evaluation. Additionally, the strong predominance of male patients in the cohort may limit the generalizability of the results to female populations.

Second, important patient-specific factors that could affect fixation stability and healing, such as bone quality and osteoporotic status, were not systematically assessed or adjusted for in our analysis.

Third, the assessment of shoulder morbidity was limited. The study did not include postoperative rotator cuff imaging to objectively evaluate soft tissue integrity. Furthermore, functional outcomes relied heavily on the Constant-Murley score. The use of other validated, patient-reported outcome measures, such as the Disabilities of the Arm, Shoulder and Hand (DASH) score or the American Shoulder and Elbow Surgeons (ASES) score, would have provided a more comprehensive and multidimensional assessment of functional recovery.

To overcome these limitations in future research, we recommend implementing routine rotator cuff imaging, incorporating a broader set of functional scores (including DASH and ASES), and systematically evaluating bone density for elderly patients. Thus, further prospective, multicenter studies with larger patient cohorts are needed to confirm and validate these findings.

Future directions

The development of third-generation nails and minimally invasive plating techniques offers promising avenues for improving outcomes in the treatment of humeral shaft fractures. Third-generation nails, which avoid tendon disruption, may reduce the risk of shoulder stiffness associated with IMN [13]. Similarly, MIPO techniques may reduce the risk of radial nerve injury and infection associated with PF [8, 30].

Conclusion

Both IMN and PF represent effective surgical options for humeral shaft fractures in older adults, though each technique carries a distinct risk–benefit profile. Intramedullary nailing is associated with shorter operative duration and a reduced risk of infection. However, it confers a greater likelihood of shoulder stiffness and persistent pain, which may adversely affect long-term functional recovery. In contrast, PF demonstrates superior fracture stabilization particularly in complex fracture patterns, and is linked with improved functional outcomes, albeit at the cost of longer surgery duration and a potentially elevated infection risk.

The better functional performance observed with PF, as measured by shoulder-specific scoring, appears attributable to two key factors: reduced postoperative pain and a lower incidence of shoulder stiffness. The antegrade nailing approach inherent to IMN necessitates rotator cuff violation, which can lead to pain and restricted motion, thereby compromising functional recovery. Plate fixation, utilizing an anterolateral approach, avoids iatrogenic injury to the rotator cuff and better preserves shoulder mechanics, contributing to more favorable functional results despite more extensive soft tissue dissection.

In light of these findings, plate fixation may be considered a preferred option for geriatric patients where optimizing functional recovery is a primary goal. Ultimately, the choice of implant should be individualized, based on fracture characteristics, bone quality, patient comorbidities, and surgical expertise. Future research should prioritize long-term evaluations of third-generation intramedullary nails designed to minimize rotator cuff disruption, as well as minimally invasive plating techniques, to further refine treatment strategies for this vulnerable population.

Authors’ contributions

B.A. and S.B. contributed to the study conception and design. A.T. and A.A. collected and analyzed the data. M.J. and B.W. performed the statistical analysis and interpreted the results. M.S.D. supervised the project and provided critical revisions. B.A. and S.B. wrote the main manuscript text. All authors reviewed and approved the final manuscript.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Review Board of the Kassab Orthopedic Institute, which granted a waiver of informed consent for participation (project reference: IMKO-CE-2023–130) based on its retrospective design.

All methods were performed in accordance with the relevant guidelines and regulations, including the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.