Outcomes of Distal Radial Fractures in Geriatric Patients Following Volar Locked versus Dorsal Bridge Plating

Abstract

Background

Volar locked plating (VLP) and dorsal bridge plating (DBP) are commonly used fixation techniques for distal radial fractures (DRFs). In patients older than 65 years, DRFs account for more than 18% of all fractures; however, a comparison of outcomes of these techniques in the geriatric population is absent.

Purpose

The purpose of our study was to use propensity score matching to compare both clinical and radiographic outcomes of VLP and DBP fixation of DRF in geriatric patients ≥65 years.

Materials and Methods

In total, 2,181 patients at least 65 years old with closed DRFs were retrospectively analyzed. Wrist range of motion (ROM) and radiographic outcome data were calculated at 6-month follow-up. Patient Reported Outcomes Measurement Information System (PROMIS) Upper Extremity (UE), Physical Function (PF), and Pain Interference (PI) were calculated at each follow-up visit. Injury characteristics, surgical complications, and patient demographic data were also analyzed. A combination of propensity score matching, multivariate analysis, t -test, and chi-square tests were used to conduct the statistical analysis.

Results

After propensity matching, a total of 1,375 patients underwent VLP, whereas 275 patients underwent DBP. Within the univariate analysis, the DBP demonstrated a higher proportion of AO Foundation/Orthopaedic Trauma Association (AO/OTA) 23-C fracture patterns, with fewer AO/OTA 23-A compared with VLP ( p  < 0.05). Multivariate analysis demonstrated greater wrist flexion and extension, radial and ulnar deviation, supination, pronation, and grip strength, and higher PROMIS PF and PI among the VLP group ( p  < 0.05). At 6-month follow-up, VLP patients demonstrated greater radial inclination (21.2 vs. 20.3 degrees), lower articular step-off (0.76 vs. 0.86 mm), with similar volar tilt and radial height ( p  > 0.05). DBP patients had higher rates of malunion (6.2 vs. 3.0%), nonunion (4.0 vs. 2.0%), and tendon rupture (2.5 vs. 0.9%) but similar rates of revision surgery and infection ( p  > 0.05).

Conclusion

VLP was associated with greater wrist flexion, extension, radial deviation, ulnar deviation, supination, pronation, grip strength, and PROMIS PF and PI compared with DBP. There were significantly higher complication rates with DBP; however, there was a similar rate of revision surgery. While the VLP group showed significantly greater ROM and radiographic outcomes, this likely did not represent a clinically significant difference in this population.

Level of Evidence

Therapeutic III.

Distal radial fractures (DRFs) are among the most common injuries in orthopaedics, affecting a wide range of age groups. ¹ The incidence of DRFs has been rising both in the United States and globally. ^{2 3 4} This rise is believed to be multifactorial, driven by factors such as increased life expectancy, a growing elderly population, changes in lifestyle habits, and higher rates of osteoporosis among older patients. ^{5 6 7 8 9} In patients older than 65 years, DRFs account for more than 18% of all fractures, and treatment of these fractures in the elderly is projected to cost about $600 million in 2025. ^{10 11} Moreover, although elderly patients typically sustain these injuries from low-energy falls that cause simple, extra-articular fractures, poor bone quality can sometimes lead to more complex, comminuted fractures. ¹²

Treatment of DRFs depends on numerous factors such as the patient's functional status, age, and fracture pattern. Many simple, extra-articular fractures can be treated with closed reduction and immobilization, whereas comminuted and intra-articular fractures, open fractures, or unstable fractures are often treated with open reduction and internal fixation (ORIF). ¹³ Despite numerous studies investigating conservative versus operative management of DRF in the elderly, controversy remains as to the best treatment option. ^{14 15 16} Most evidence indicates that long-term (>1 year), patient-reported outcomes are comparable between the two treatments, regardless of radiographic outcomes. ¹⁷ Nevertheless, short-term outcomes with operative fixation may be favorable with some studies demonstrating that ORIF with plate fixation, allowing for earlier functional recovery and mobility in elderly patients. ^{17 18} Additionally, locking plate fixation may allow for fewer postoperative radiographs for monitoring. ¹⁹ Despite the lack of consensus, there has been a trend toward operative management of DRFs in elderly patients, specifically with ORIF, possibly due to the increasing life expectancy and more active lifestyles of older patients, and desire for quicker recovery. Elderly patients are now having longer disability-free life expectancies, challenging the traditional belief that these patients have low functional demands. ^{20 21} Patients, even of the same age, can have substantially different activity levels and, therefore, some geriatric patients may benefit from the improved short-term outcomes of operative fixation. ^{22 23 24}

The two commonly used ORIF techniques for DRFs include volar locking plating (VLP) and dorsal bridge plating (DBP). VLP is the most common fixation method and can be used for a wide variety of DRFs, allowing for early range of motion (ROM). ²⁵ DBP, in contrast, has been historically utilized for comminuted distal fracture patterns and polytraumatized patients, similar to wrist-spanning external fixators, which are associated with complications such as pin-site infections and stiffness. ^{26 27 28} Although it does require a second procedure for hardware removal, DBP may confer immediate biomechanical stability and allow for early partial weight-bearing, which can be particularly useful in elderly patients who require assistive devices for ambulation. ^{29 30 31}

Most available studies in the elderly population specifically investigate the outcomes of VLP, with a notable paucity of literature comparing the outcomes of VLP and DBP in elderly patients. The purpose of our study was to use propensity score matching to compare both clinical and radiographic outcomes of VLP and DBP fixation of DRF in geriatric patients (≥65 years).

Materials and Methods

This was a propensity score-matched retrospective evaluation of a single Level 1 trauma center. The study was approved by our institutional review board (IRB) and a waiver of consent was granted. Patients included in this study were identified utilizing Current Procedural Terminology codes 25607, 25608, and 25609 and operative technique was determined via chart review. Inclusion criteria were patient ages 65 to 90 years old, who underwent operative treatment of DRFs between October 1, 2015 and October 1, 2023. Exclusion criteria included cases of revision surgery, ages outside the above range, polytraumatized patients, previous wrist surgery, and patients without at least 6 months of follow-up. The data were deidentified and securely stored within the hospital network. Following exclusion, 348 DBP and 1,833 VLP patients were eligible for matching.

Patients meeting the specified inclusion and exclusion criteria were matched using a propensity score methodology at a ratio of 1:5. Propensity scores were generated through logistic regression modeling, accounting for age, body mass index (BMI), sex, race, AO Foundation/Orthopaedic Trauma Association (AO/OTA) classification (23 A, B, or C), and comorbidities like tobacco use, diabetes, hypothyroidism, and rheumatoid arthritis, which were considered as covariates. BMI categories were defined as less than 20, 20 to 25, 25 to 30, 30 to 35, 35 to 40, and greater than 40 kg/m ² . Race was categorized as Black or African American, White, or other. Matching was conducted using a greedy, nearest-neighbor algorithm, with a 1:5 ratio of DBP to VLP, and without replacement. To ensure high-quality matches, a caliper was implemented to limit the distance between propensity scores. The caliper was set at 0.2 times the standard deviation of the logit of the propensity scores for the entire cohort. ^{32 33} Following this matching procedure, the resulting groups comprised 275 DBP patients and 1,375 VLP patients.

The surgical procedures were generally conducted by either a hand surgeon or an orthopaedic trauma surgeon, with a small percentage of cases were managed by on-call personnel at a tertiary Level 1 trauma center. A total of 18 surgeons participated in the study utilizing a variety of implant designs for volar locked plate (VLP) and dorsal bridge fixation. Patients undergoing DBP in this study underwent closed reduction and fixation to the 2nd or 3rd metacarpal shaft, based on the surgeon's preference. Postoperatively, all patients were immobilized in a volar plaster splint for 2-4 weeks, followed by a removable wrist brace. Plate removal for the DBP group was typically conducted as part of standard practice approximately 2-3 months post-surgery, at the surgeon's discretion. Rehabilitation for the VLP cohort typically commenced between 1 and 6 weeks postoperatively, as determined by the surgeon, while therapy for the dorsal bridge plate (DBP) cohort began after plate removal. Patients were instructed to initiate finger ROM exercises at home and to elevate the affected limb for edema control. Subsequently, all patients followed similar postoperative rehabilitation protocols under the guidance of hand therapists.

Pre- and postoperative plain radiographs were evaluated independently by three orthopaedic surgeons, blinded to the patients' identities when assessing the preoperative radiographs, even though fixation technique was apparent on postoperative imaging. Only plain films were examined even when advanced imaging was available. Preoperative AO/OTA classification was determined, and patients were categorized as AO/OTA 23-A, 23-B, or 23-C. Radial height, radial inclination, and articular step-off were measured from posteroanterior (PA) radiographs, whereas volar tilt was calculated from lateral radiographs. Both lateral and PA radiographs were obtainable for all patients enrolled in the study.

Patient demographics (such as age, sex, BMI, and race) as well as comorbidities (tobacco use, diabetes, hypothyroidism, and rheumatoid arthritis) were obtained through chart review. Patient-reported outcome measures (PROMIS PF v1.2/2.0, UE v2.0, and PI v1.1) were collected using computer adaptive tests during routine clinic visits conducted between October 1, 2015, and October 1, 2023, utilizing iPad tablets. Patient Reported Outcomes Measurement Information System (PROMIS) scores were routinely obtained at the postoperative 6-month follow-up, with a final completion percentage of 76%. Surgical complications, including the need for revision surgery, malunion, nonunion, superficial or deep infection, and tendon rupture, were determined through chart review of clinical notes and operative reports. Asymptomatic and symptomatic nonunion/malunion cases were combined for the study's purposes, with malunion defined as deviation from the established acceptable radiographic parameters.

ROM and strength measurements, including wrist flexion, wrist extension, forearm pronation, forearm supination, radial deviation, ulnar deviation, and grip strength, were assessed. These measurements were conducted collaboratively by dedicated hand therapists and participating hand surgeons utilizing a goniometer and calibrated hydraulic hand dynamometer. Assessment of ROM and grip strength occurred at the 6-month mark following the final surgical procedure, including the index procedure for the VLP group and the DBP removal date for the DBP group. Additionally, measurements were taken at 6 months following the initial surgical procedure, irrespective of the plate removal date.

Descriptive statistics, presented as mean and standard deviation, were utilized to summarize the data. Univariate analysis was employed to compare baseline demographics and fracture classifications, as well as to assess matching balance between groups. Clinical outcomes were examined using a combination of chi-square tests, pooled and unpooled t -tests, and two-proportion z-tests. Standardized mean differences were employed to compare matched cohort characteristics. Differences in follow-up duration were adjusted for using mixed effects regression modeling. A significance level of 0.05 was set for all statistical tests. Statistical analysis was performed using the R statistical software (version 4.3.0; R Foundation for Statistical Computing).

Results

After exclusion criteria, a total of 2,181 patients were included in this study, with 1,833 patients in the VLP group and 348 in the DBP group ( Fig. 1 ). Baseline characteristics among the unmatched cohort in both groups can be found in Table 1 . Patients who underwent DBP were significantly older than those with VLP (78.4 vs. 73.9 years; p  < 0.05). The DBP cohort also had a higher proportion of patients who were smokers and diabetics ( p  < 0.05). Additionally, the DBP group was less likely to include AO/OTA 23A DRFs and more likely to include 23C patterns ( p  < 0.05).

Fig. 1.

Flow Diagram illustrating geriatric patients who underwent volar locked plating or dorsal bridge plating included in this study.

Table 1. Unmatched cohort characteristics.

Characteristic	Volar locked plating ( n  = 1,833)	Dorsal bridge plating ( n  = 348)	p -Value
Demographics
Age (SD)	73.9 (7.0)	78.4 (6.8)	0.0001
BMI > 40 kg/m 2 (%)	37 (2)	3 (1)	0.32
Female (%)	1,081 (59)	223 (64)	0.12
Race (%)
Black or African American	110 (6)	21 (6)	0.89
White	1,613 (88)	299 (86)	0.41
Other	110 (6)	28 (8)	0.32
Comorbidities (%)
Tobacco use	458 (25)	104 (42)	0.04
Diabetes	202 (11)	51 (15)	0.04
Rheumatoid arthritis	111 (6)	14 (4)	0.20
Hypothyroidism	92 (5)	17 (5)	0.96
AO/OTA classification (%)
23-A	202 (11)	7 (2)	0.0001
23-B	697 (38)	121 (35)	0.25
23-C	934 (51)	220 (63)	0.0001

Abbreviations: AO/OTA, AO Foundation/Orthopaedic Trauma Association; BMI, body mass index; SD, standard deviation.

After 5:1 propensity matching, 1,375 VLP patients were matched with 275 DBP patients, for a total of 1,650 patients. There were no significant differences between the cohorts and the matches were balanced in all variables examined ( Table 2 ).

Table 2. Propensity matched cohort characteristics.

Characteristic	Volar locked plating ( n  = 1375)	Dorsal bridge plating ( n  = 275)	Standardized difference
Demographics
Age (SD)	76.8 (6.2)	77.1 (6.1)	0.04
BMI > 40 kg/m 2 (%)	15 (1)	3 (1)	0
Female (%)	839 (61)	171 (62)	0.03
Race (%)
Black or African American	105 (8)	21 (8)	0
White	1201 (88)	237 (86)	0.04
Other	69 (5)	17 (6)	0.05
Comorbidities (%)
Tobacco use	344 (25)	72 (26)	0.03
Diabetes	141 (10)	25 (9)	0.04
Rheumatoid arthritis	68 (5)	11 (4)	0.03
Hypothyroidism	69 (5)	14 (5)	0.01
AO/OTA classification (%)
23-A	151 (11)	18 (2)	0
23-B	522 (38)	29 (35)	0
23-C	701 (51)	56 (63)	0

Abbreviations: AO/OTA, AO Foundation/Orthopaedic Trauma Association; BMI, body mass index; SD, standard deviation.

Outcomes analysis performed between the groups at 6-month follow-up demonstrated improved ROM outcomes (wrist flexion/extension, forearm pronation/supination, and radial/ ulnar deviation) in VLP as compared with DBP patients ( p  < 0.05). Grip strength, as measured by percent comparison with the uninjured side, was also significantly greater in the VLP group ( p  < 0.05). PROMIS physical function (PF) and pain interference (PI) scores were worse in DBP patients ( p  < 0.05), whereas upper extremity (UE) scores were similar ( Table 3 ).

Table 3. Propensity matched range of motion and patient-reported outcomes at 6-month follow-up.

Outcome	Volar locked plating ( n  = 1,375)	Dorsal bridge plating ( n  = 275)	p -Value
Range of motion (degrees; SD)
Wrist flexion	61.6 (4.8)	57.3 (6.0)	0.0001
Wrist extension	65.0 (5.8)	58.6 (4.2)	0.0001
Forearm pronation	83.7 (1.4)	82.4 (1.6)	0.0001
Forearm supination	82.9 (1.0)	80.8 (1.3)	0.0001
Wrist radial deviation	17.6 (2.1)	17.3 (2.4)	0.035
Wrist ulnar deviation	24.6 (2.4)	24.2 (2.7)	0.014
Grip strength (% uninjured)	68.9 (14.0)	60.3(13.6)	0.0001
Patient-reported outcomes (SD)
PROMIS UE	33.9 (5.0)	33.6 (3.5)	0.34
PROMIS PF	45.8 (3.2)	44.8 (3.7)	0.0001
PROMIS PI	52.4 (3.9)	51.8 (4.2)	0.022
Radiographic outcomes (SD)
Radial height (mm)	11.6 (1.6)	11.4 (2.0)	0.07
Radial inclination (degrees)	21.2 (3.5)	20.3 (3.7)	0.0001
Volar tilt (degrees)	6.9 (1.2)	7.0 (1.2)	0.21
Articular step-off (mm)	0.76 (0.2)	0.86 (0.2)	0.0001

Abbreviations: PF, Physical Function; PI, Pain Interference; PROMIS, Patient Reported Outcomes Measurement Information System; UE, Upper Extremity; SD, standard deviation.

With respect to radiographic outcomes, even though both groups were within acceptable limits (<2-mm articular step-off, <5-degree dorsal tilt, <5-degree change in radial inclination, and <5mm radial shortening), VLP patients had more significant improvements in radial inclination (21.2 vs. 20.3 degrees, p  < 0.05) and articular step-off (0.76 vs. 0.86, p  < 0.05). There was no difference in measurements of radial height and volar tilt between the groups ( Table 3 ).

Complication rates remained overall low in both groups, but the VLP patients did have significantly fewer rates of malunion (3 vs. 6%, p  < 0.05), nonunion (2 vs. 4%, p  < 0.05), and tendon rupture (1 vs. 3%, p  < 0.05). Despite this, surgical revision rates were similar between the VLP and DBP groups (5 vs. 4%, p  > 0.05; Table 4 ).

Table 4. Propensity matched complication rates at 6-month follow-up.

Outcome	Volar locked plating ( n  = 1,375)	Dorsal bridge plating ( n  = 275)	p -Value
Complications
Need for revision surgery	69 (5)	11 (4)	0.47
Malunion	41 (3)	17 (6)	0.008
Nonunion	27 (2)	11 (4)	0.040
Superficial infection	41 (3)	6 (2)	0.47
Deep infection	14 (1)	3 (1)	0.91
Tendon rupture	13 (1)	7 (3)	0.027

Discussion

Our study shows an improvement in outcomes in geriatric patients (>65 years old) who underwent VLP compared with those with DBP. The VLP cohort demonstrated greater wrist flexion and extension, radial and ulnar deviation, supination, pronation, grip strength, and PROMIS PF and PI, as well as improved radiographic parameters. This group also showed lower rates of malunion, nonunion, and tendon rupture. Nevertheless, these differences, even although they were significant, they might not be clinically relevant.

VLP is the most common method of fixation of DRFs, regardless of the patient's age. It allows for fixation of a variety of fracture patterns with good clinical outcomes. ²⁵ DBP, on the other hand, has emerged as a fixation method that is optimal for severely comminuted fractures that are difficult to stabilize with a VLP as was exemplified in our patient population, where a higher proportion of AO/OTA 23-C fracture patterns were treated with DBP compared with VLP. Nevertheless, in the elderly population specifically, nonoperative treatment is often favored, as studies have shown that surgery may not offer superior long-term clinical outcomes. ^{34 35} Surgical treatment, however, and specifically DBP, allows for early partial weight-bearing which can be very beneficial in elderly patients that rely on assistive devices for ambulation and may allow for faster return to previous activity level. ^{29 30 31 36}

We found higher PROMIS PF and PI in the VLP group compared with DBP in our patient cohort. Some studies have investigated patient-reported outcomes of VLP compared with external fixation, percutaneous pinning, and nail fixation, and, to our knowledge, only one other study has looked at PROMIS scores after DRF surgery. ³⁷ Hollenberg et al, investigated PROMIS outcomes in younger and older patients following DRF surgery and found a significant improvement in both PROMIS PF and PI scores for older patients at a 6-month follow-up. However, the extent of improvement was notably lower than in younger patients, and the study exclusively included VLP-treated patients. ³⁸ Another relevant study by Carroll et al examined outcomes for VLP versus DBP across a broader age range. Similar to our findings, they reported an improvement in PROMIS PF with VLP but found no significant difference in PROMIS PI between the two groups. ³⁹ The minimal clinically important difference (MCID) for PROMIS PF and PI after DRFs have been reported to be 5.2 and 6.8, respectively, and it is worth noting that while our cohort shows higher PROMIS PF and PI scores in the VLP group compared with the DBP group, the differences did not reach these MCID thresholds. ⁴⁰

In our geriatric population, VLP demonstrated greater wrist flexion and extension, radial and ulnar deviation, supination, pronation, and grip strength. These improvements were also noted by Carroll et al, but other studies have not made a direct comparison between the ROM and grip strength of patients undergoing VLP versus DBP. ³⁹ Our results overall align with this study, demonstrating superior ROM improvements with VLP as well. Even though the MCID for ROM and grip strength after DRFs have not been established in the literature, patient dissatisfaction has been noted with 65% grip strength and 79 to 95% of wrist ROM compared with the uninjured sides. ^{41 42} In the VLP group, average wrist ROM was 126.6 degrees and grip strength was 68.9% compared with the contralateral side, whereas DBP average wrist ROM was 115.9 degrees with grip strength of 60.3%. Using the reported criteria, wrist ROM in both groups appear to be within the acceptable, satisfactory parameters but grip strength in the DBP was less than 65% and may have clinical relevance in patient dissatisfaction.

VLP fixation has been shown to achieve significant improvements in key radiographic parameters, including dorsal tilt, radial inclination, and radial shortening, when compared with K-wires. Additionally, VLP has demonstrated superior control over ulnar variance and radial inclination in comparison to external fixation. ^{43 44} Interestingly, one study demonstrated a significant correlation between ulnar variance and QuickDASH and Mayo scores and grip strength. ⁴⁵ Our study did not look specifically at ulnar variance, however, but this may be an area for further investigation. In our analysis, we found that VLP had superior radial inclination and lower articular step-off, but similar volar tilt and radial height compared with DBP. Schmidt et al reported that dorsal tilt was the radiographic parameter that has the strongest and nonlinear association with patient-reported outcomes, grip strength, and ROM. ⁴⁶ They established that even though a decline in clinical outcomes starts to occur at 5 degrees of dorsal tilt, patients are unlikely to notice a difference until approximately 20 degrees. ⁴⁶ Bearing this in mind, the improvements noted in our radiographic parameters are minimal and may not be substantial enough to influence clinical outcomes between fixation methods.

In our cohort, complication rates, including malunion, nonunion, and tendon rupture, were lower in the VLP group compared with the DBP group, aligning with previous literature that reports VLP complication rates between 6 and 30% and DBP rates around 13%. ^{47 48 49} Townsley et al noted that early complications (<90 days) after DRF surgical fixation include tendon damage or rupture, stiffness, and complex regional pain syndrome, whereas late complications (>90 days) can be symptomatic hardware, distal radioulnar joint instability, arthritis, malunion, and further tendon complications. ⁵⁰ In our study population, while union and tendon rupture rates were higher with VLP, there was no significant difference in infection or revision surgery rates between VLP and DBP. Similar revision rates may be due to patients remaining asymptomatic or satisfied with their function, even when complications were present, reducing the necessity for further surgical intervention.

A limitation of our study is the nonrandomized selection process for VLP versus DBP, with treatment decisions made collaboratively by patients and surgeons. This approach may have introduced selection bias, as certain patient characteristics and potentially cost considerations could have influenced the choice of fixation method. Even though baseline demographics between the two groups differed in terms of age, tobacco use, diabetes status, and fracture OTA classification, we applied propensity matching to account for these variables. Notably, while an increased BMI may be associated with more severe DRFs, our study did not demonstrate a significant difference in the proportion of patients with a BMI greater than 40 in the two cohorts. ⁵¹ Additionally, polytrauma patients were excluded from our analysis, despite DBP often being indicated in such cases. Including these patients could have impacted outcomes, as they commonly have concurrent upper and lower extremity injuries that could affect recovery and functional results. Regarding study outcomes, radiographic assessments were conducted by three physicians; however, interrater reliability was not evaluated, which could impact the consistency of these findings. ROM and strength data were obtained through chart review without independent verification, raising potential concerns about measurement accuracy. Additionally, our study's follow-up period was relatively short at 6 months postoperatively, and clinical and radiographic outcomes between VLP and DBP may converge over a longer time.

Despite these limitations, our study benefits from a large sample size and is the first to directly compare outcomes of VLP and DBP outcomes in a geriatric population, thereby contributing valuable insight to the existing literature. Our findings suggest that while VLP may offer some statistically significant advantages over DBP, both remain viable surgical options for the treatment of DRFs in elderly patients.