Intra-Articular Antegrade Intramedullary Screw Fixation for Proximal Phalanx Fractures: Impact of Articular Surface Defects on Joint Contact Pressures

Abstract

Background:

Intramedullary headless screw fixation is increasingly used for fixation of proximal phalanx fractures. However, the impact of screw entry defects on joint contact pressures is not well defined and may have implications for arthrosis. The objective of this cadaveric biomechanical study was to assess joint contact pressures at the metacarpophalangeal (MCP) joint before and after passage of 2 sizes of antegrade intramedullary fixation.

Methods:

Seven fresh frozen cadaver specimens without arthritis or deformity were included in this study. Antegrade intramedullary screw fixation of proximal phalanx fracture was simulated using an intra-articular technique. Flexible pressure sensors were inserted into the MCP joints and cyclic loading was performed. Peak contact pressures were determined and averaged across loading cycles for each finger in the native state, with 2.4- and 3.5-mm drill defects in line with the medullary canal.

Results:

Peak pressure increased with the size of the drill hole defect. Contact pressure increases were greater in extension, with peak contact pressures increased by 24% for the 2.4-mm defect and 52% for the 3.5-mm defect. Increase in peak contact pressure was statistically significant with a 3.5-mm articular defect. Contact pressures were not consistently increased for the 2.4-mm defect. Testing in flexion of 45° resulted in reduced contact pressure for these defects.

Conclusions:

Our study demonstrates that antegrade intramedullary fixation of proximal phalanx fractures can increase MCP joint peak contact pressures, particularly in an extended joint position. Effect increases with defect size. This has implications for the management of proximal phalanx fractures using this technique.

Introduction

The primary goal of treating unstable proximal phalanx fractures is to achieve alignment and stability of the fracture fragments to promote healing while minimizing scar and contracture. One increasingly popular technique used to achieve this is intramedullary headless screw fixation, which provides biomechanical stability similar to that of plate and screw constructs. ¹ This provides rigid fixation to allow for early active motion, while causing minimal or no periosteal dissection or hardware prominence.^1-8 However, creating a cortical defect at the point of screw insertion could result in altered joint contact pressures and later arthrosis.^4,6,8

While current clinical studies of intramedullary fixation suggest positive early results, these studies are limited by the duration of follow-up.^2-4, ⁷ Imaging studies have been carried out to estimate the effect of circular defects on articular surface area involvement and suggest that the articular loss as a percentage of total articular surface area is low. ⁴ ,^6-9 However, using the total articular surface area for the calculation to estimate changes in contact pressure may be inaccurate as distribution of cartilage contact stress area varies during motion. ¹⁰ A direct measurement of contact pressures at the joint surface before and after creation of an articular defect may better represent changes in joint dynamics in vivo.

The purpose of this study was to measure joint contact pressures at the metacarpophalangeal (MCP) joint before and after passage of 2 sizes of antegrade intramedullary headless compression screw into the base of the proximal phalanx in a cadaveric biomechanical model. Our hypothesis was that peak joint contact pressures would increase with creation of a defect in the articular surface and increase further depending on the size of the defect.

Materials and Methods

This study was approved by the Institutional Biosafety Committee. This is a biomechanical cadaveric study. Seven cadaveric arm specimens were included from 4 male and 3 female donors with a mean age of 50.2 years (range: 35-56) and mean body mass of 73.6 kg (range: 54.1-93.6). Specimens had not undergone previous forearm or hand surgery and were evaluated radiographically to confirm the lack of deformity or evidence of arthritis before use. The specimens were kept at a temperature of −17°C until they were needed for testing, at which point they were warmed to room temperature.

Specimen Preparation

The cadaver specimen was prepared by sectioning at the mid forearm and removal of overlying skin from the MCP joints to the forearm. The extensor apparatus overlying each MCP joint was incised along the ulnar aspect and retracted radially to reveal the joint capsule. This was incised transversely to visualize the joint. A volar approach was made after similar removal of skin palmarly, and the A1 pulley was incised and flexor tendons retracted to reveal the volar plate. The volar plate was released from its proximal attachments at the metacarpal neck and retracted distally to visualize the joint.

For each specimen, calibrated flexible pressure sensor arrays (Model 6900; Tekscan Inc, Boston, Massachusetts) with an active sensing area of 14 mm × 14 mm and a spatial resolution of 62 sensors per cm² were placed in the joint space covering the base of the proximal phalanx (Figure 1). Sensors were inserted dorsal to volar and secured to the volar plate to provide stability to the sensor in relation to the proximal phalangeal base. Before calibration, polyvinyl chloride adhesive tape was folded over the sensors with a suture adhered inside the tape fold. This created a cuff for securing the sensor array and calibration of the sensor, and this was sutured to the volar plate tissue.

Figure 1.

Prepared cadaveric specimen with secured sensor arrays and transverse metacarpal screws for mounting specimen to testing apparatus.

The flexor and extensor tendons were dissected independently at the dorsal and volar forearm. Using a free needle, high test braided line was sutured through the flexor and extensor tendons in Krakow fashion with long tails for later attachment to the testing apparatus (Figure 1).

Equipment

The test specimen was mounted securely to a fixed test bed via transverse M4 screws through the metacarpals. Flexor and extensor tendons were connected to a rigid bracket via spring force gauges. These gauges could be individually adjusted to apply a balanced force across each MCP joint in varying degrees of flexion. The bracket was mounted on a motorized linear slide that allowed the wrist to be cyclically offloaded and loaded to the preset force levels (Figure 2).

Figure 2.

Testing apparatus with rigid bracket (left) to secure the mounting screws in the metacarpals, and spring force gauges (right) to individually adjust and apply a balanced force across each metacarpophalangeal (MCP) joint in flexion and extension. The linear slide was adjusted to allow for the finger to be cyclically loaded to the preset force levels with a motorized driver.

Test Protocol

Initially, each joint was tested without any articular defect (control condition). The flexor and extensor tendons were loaded such that a combined load of 50 N was placed across the joint in neutral and 45° flexion positions. These joint positions simulate many activities of daily living. A 90° testing position was considered, but after dorsal and volar arthrotomy for placement of pressure sensors and A1 pulley release, the vector of pull of the extrinsic flexors and extensors tended to subluxate the proximal phalanx proximally, and we did not feel that this simulated physiological contact stresses on the MCP joint.

A combined load of 50 N was selected as a joint load in the physiological range. Hu et al ¹¹ estimated that isometric fingertip pressing load from a representative sample was estimated in the range of 0 to 30 N. They calculated MCP bone-to-bone contact forces using models and estimated they are in the range of 10.5 to 13.0 times applied fingertip load for fingers. ¹¹ Therefore, MCP joint contact forces are up to 390 N in healthy 25-year-old adults. ¹¹ We felt 50 N was a physiologically realistic force on the joint and adequate force to note changes in contact pressures.

In each position, the MCP joints underwent 10 cycles of cyclic loading at a frequency of 0.2 Hz. Pressure data were sampled at a rate of 100 Hz during the loading cycles. The proximal and distal interphalangeal joints were splinted in extension for consistent application of force at the MCP joint. After testing both joint positions, an articular defect was created in the proximal phalanx to mimic the defect resulting from drilling a hole using an intra-articular technique for an intramedullary screw at the dorsal central aspect of the bone (Figure 3). In the first of these articular defect conditions, a 2.4-mm drill bit was used to create the defect. Cyclic loading with pressure measurements was repeated and recorded at both joint angles identical to the control condition. Once complete, the previously created articular defect was then expanded to 3.5 mm with an appropriate drill bit, and the loading cycles repeated. Therefore, each joint included was tested at 2 positions for 3 test conditions.

Figure 3.

Circular articular defect in dorsal central base of the proximal phalanx articular surface. This aligns with the medullary canal of the phalanx.

Data Processing and Analysis

Peak contact pressure was determined and averaged across loading cycles for each position/condition. A mixed effects regression model was used to determine whether the fixation hole had a significant effect on contact pressures. The pressure measurements were modeled as the dependent variable, the surgical condition as the fixed effect, and to account for the repeated measurements taken from each specimen, individual specimens as random effects. A value of P = .05 was considered significant. If a significant association was found, pairwise comparisons were performed with the Tukey range test used to correct for multiple comparisons. Pairwise results are presented as adjusted P values, meaning that all results can be interpreted as statistically significant if below .05.

Results

A total of 25 MCP joints were included in the analysis (6 index, 6 middle, 7 ring, and 6 little). Three fingers were excluded due to technical issues including joint instability and damage to the pressure sensor.

Contact pressure maps were generated for the native condition, 2.4-mm hole, and 3.5-mm hole. In the native condition, peak contact pressures were least for the middle finger and greatest for the small finger (Table 1). The drill hole conditions demonstrated central defects on contact pressure maps consistent with the articular defect (Figure 4). Peak pressure was generally noted to increase with increasing size of the screw hole, and in no circumstances did a drill hole result in reduced peak contact pressure (Table 1).

Table 1.

Pressure Measurements Across Test Conditions for the Metacarpophalangeal (MCP) Joint of the Second Through Fifth Digits in N/mm².

	Peak pressure in N/mm2 (±SD)
Digit and position	Native condition	2.4-mm hole	3.5-mm hole
MCP2 extended	2.7 (±0.7)	3.7 (±1.2)*	4.0 (±1.3)*
MCP2 flexed 45°	3.1 (±0.9)	3.1 (±0.8)	4.6 (±1.4)**
MCP3 extended	2.1 (±0.6)	2.7 (±0.6)*	3.1 (±0.4)**
MCP3 flexed 45°	2.6 (±0.8)	2.6 (±0.6)	3.0 (±1.1)
MCP4 extended	2.3 (±0.7)	2.8 (±0.7)	3.7 (±1.2)*
MCP4 flexed 45°	2.5 (±0.8)	2.8 (±0.8)	3.7 (±1.8)**
MCP5 extended	3.4 (±1.2)	3.7 (±0.8)	5.1 (±2.0)*
MCP5 flexed 45°	3.2 (±0.8)	4.6 (±2.2)	4.6 (±1.6)

^*Significant difference compared with native condition. **Significant difference compared with native condition and 2.4-mm hole.

Figure 4.

Example pressure maps for 1 hand across various conditions.

The 2.4-mm articular defect showed a mean increase peak pressure of the MCP joint of 24% in the extended position, but there was no significant difference in the 45° flexion position. The 2.4-mm hole showed no significant difference in the ring or small finger in extension or flexion (Table 1). The 3.5-mm articular defect showed a mean increase in peak pressure of the MCP joint of 52% across all joints in the extended position. Statistical significance was seen when comparing the 3.5-mm articular defect with the native condition for all joints in the extended position (Tables 1 and 2). In the 45° flexed position, peak contact pressure increased from 2.6 to 3.0 N/mm² for the middle finger, and from 3.2 to 4.6 N/mm² but this was not statistically significant in our study.

Table 2.

Results of Statistical Comparisons With P Values.

Joint position	Native vs 2.4-mm hole	Native vs 3.5-mm hole	2.4-mm vs 3.5-mm hole
MCP2 extended	P = .003 (0.3, 1.7)	P < .001 (0.6, 2.0)	P = .49 (−0.3, 1.1)
MCP2 flexed	P = .99 (−0.9, 1.0)	P = .001 (0.5, 2.5)	P = .001 (0.6, 2.5)
MCP3 extended	P = .002 (0.2, 0.9)	P < .001 (0.6, 1.4)	P = .02 (0.1, 0.8)
MCP3 flexed	P = .99 (−0.6, 0.6)	P = .35 (−0.2, 1.0)	P = .33 (−0.2, 1.0)
MCP4 extended	P = .40 (−0.4, 1.4)	P = .002 (0.4, 2.2)	P = .06 (−0.1, 1.7)
MCP4 flexed	P = .68 (−0.5, 1.1)	P = .002 (0.4, 2.0)	P = .02 (0.1 1.7)
MCP5 extended	P = .91 (−1.3, 1.9)	P = .04 (0.1, 3.3)	P = .14 (−0.3, 3.0)
MCP5 flexed	P = .20 (−0.5, 3.4)	P = .30 (−0.8, 3.4)	P = .98 (−1.9, 2.2)

Note. Bracketed values indicate 95% confidence intervals. MCP = metacarpophalangeal.

Discussion

In this study, we compared the peak contact pressures of the MCP joint for the native joint and 2 diameters of drill holes typically used for placement of antegrade intramedullary headless screw fixation in the proximal phalanx. We simulated an intra-articular technique. We found that peak contact pressures were increased by drill holes and that peak contact pressures were consistently and significantly increased by the 3.5-mm drill hole for all digits in the extended testing position. The 2.4-mm drill hole defect change was statistically significant in the index and middle finger MCP joints in the extended testing position, but was not otherwise statistically significant.

The clinical significance of increased peak contact forces is not yet fully understood, but extended periods of supraphysiological stress on chondrocytes have been associated with chondrocyte death and extracellular matrix damage. ¹² A contact pressure greater than 15 N/mm² is associated with chondrocyte death and extracellular matrix damage.^12,13

Our testing conditions of 50-N force did not cause supraphysiological peak pressures, but our model is likely in a low physiological range appropriate for daily activities. Estimation of typical physiological loads suggests that an external fingertip force of 46.7 ± 10.3 N at the index finger tip would result in an MCP joint reaction force of 585.9 ± 119.8 N, or ~12.5 times the fingertip force. ¹³ During a maximal grip test, mean contact forces for the index finger MCP joint are estimated to be 7.2 ± 1.3 N/mm². ¹³ Based on these estimates, and the relative increase in peak contact pressures of measured in extension positions, it is possible that defects of increasing size may result in supraphysiological loads on chondrocytes with higher demand activities.

Previous studies have suggested that articular defects from placement of intramedullary headless screw fixation are likely negligible as they represent a small portion of the total articular surface. ⁴ del Piñal et al ⁴ have estimated the impact of defects using computed tomography scans of healthy digits. Borbas et al ⁶ performed a cadaveric study looking at the articular defects with 2 screw sizes, 2.2 and 3.0 mm, and made estimates of the impact of defects using standardized digital photographs of the base of the proximal phalanx and measurements obtained by 2 observers using image processing software. The limitation of imaging estimates of total surface area is that different areas of cartilage have greater or lesser contact during motion. ¹⁰ For the MCP joint, the main point of contact is near the dorsal base of the proximal phalanx in extension and the palmar base of the proximal phalanx in flexion. ¹⁰ This would suggest that in full flexion power grip, there may be less peak contact pressure around a dorsal central chondral lesion than in extension.^2,10 Drill holes at the outer margin of the articular surface may have little or no influence joint contact forces during physiological motion, but this has not been specifically studied.

Long-term evaluations of the significance of articular defects in this critical contact area are lacking. Although it is worth noting that scaphoid fracture fixation is accomplished in a similar intra-articular fashion, and mid-term follow-up studies have shown no significant arthritis,^14,15 there are other factors that could contribute to the clinical relevance of intramedullary screw fixation in the scaphoid. Rikli et al demonstrated using in vivo pressure sensors that during physiological movement of the wrist, more force is transmitted across the lunate and the ulnar side of the wrist joint than the scaphoid, and the total contact surface area of the wrist is larger. The MCP joint loads and kinematics are different.

The main limitation of this study is that as a biomechanical study, only the mechanical influences on contact forces are evaluated. Biological factors such bone healing and formation of fibrocartilage may influence joint contact forces. There may also be other impacts of articular defects on synovial fluid dynamics, inflammatory mediators, and cells within the joint that could influence long-term joint health. We also note that this cadaveric testing rig also has some limitations. Neutral and 45° of flexion were selected as 2 positions within a physiological range, but additional testing with the MCP joint in 90° of flexion would provide additional information into the research question. Owing to limitations of this testing rig and soft tissue releases required for sensor insertion, it may be difficult to accurately achieve balanced 50 N of joint contact force at 90° using tension on the forearm flexors and extensors, so this may be better achieved with a robotic testing rig. A model that incorporates factors related to in vivo physiological forces could also alter results. ¹⁶

In summary, we conclude that MCP joint peak contact pressures are increased after creation of dorsal central defects simulating antegrade intramedullary fixation of the proximal phalanx using an intra-articular technique. Contact pressures increase with the size of the defect and are observed to be greater in an extended joint position. Peak contact forces were not supraphysiological under the testing conditions used here (50 N), and an antegrade 2.4-mm cannulated headless screw through the MCP joint likely causes little increase in peak contact forces at the MCP joint. However, intramedullary fixation of the proximal phalanx using a 3.5-mm defect or larger could result in supraphysiological peak pressures under heavy loads.