Excision of the Distal Pole of the Scaphoid and the Midcarpal Joint

Abstract

Background Excision of the distal pole of the scaphoid is used to treat arthritis of the scaphotrapezial trapezoid (STT), radioscaphoid joint, and arthritis following scaphoid nonunion. Some patients develop midcarpal instability limiting utilization of this technique. Why some wrists develop postoperative instability while others do not, remains unclear.

Purpose To identify the wrists prone to developing midcarpal joint instability we evaluated the effect of midcarpal joint structure on force transfer through the wrist, we hypothesized that the force transfer will be further altered when a distal pole excision is performed and that midcarpal joint structure will affect force transfer.

Materials and Methods We used finite element analysis based on 19 computer tomography wrist scans. Nine type 1 (lunate has a facet with the capitate alone) and 10 type 2 (lunate has facets with both the capitate and hamate) models were prepared. A 200 N force was evenly split and applied to the dorsal crests of the trapezoid and capitate (100 N along each crest) to replicate the performance of a knuckle push-up. Displacement of the trapezoid, trapezium, scaphoid, capitate, and hamate was measured along each axis after the applied load. The simulation model was used to predict motion at the capitate and STT joint with excision of the distal pole.

Results Excision of the distal pole of the scaphoid affected the transfer of forces significantly (∼200% all bones in all directions) in all wrists. There are significant differences in force transfer between type 1 and type 2 wrists in the amount of force transferred (type 1 > type 2), in the percent difference from an intact wrist (type 1 > type 2) and in the direction of displacement (type 1 the bones moved in different directions while type 2 wrists moved as one block).

Conclusion This study suggests that midcarpal joint structure affects force transfer through the wrist and may predict wrist behavior following excision of the distal pole of the scaphoid. Specifically, type 1 wrists may be more prone to midcarpal joint collapse after excision.

Level of Evidence: Level 1.

Keywords

instability - lunate type - midcarpal joint - scaphoid - scaphoidectomy - wrist

Partial scaphoid excision has been employed to treat arthritis of the scaphotrapezial trapezoid (STT) and radioscaphoid joint as well as arthritis following scaphoid nonunion.[1] [2] [3] [4] Because a proportion of the patients develop midcarpal instability with or without clinical symptoms, utilization of this technique has been limited, and it is used in many cases as an adjunct to an intercarpal partial fusion procedure that provides midcarpal joint stability, such as radioscaphocapitate fusion.[5] [6] [7] While in some cases midcarpal joint instability is symptomatic and can develop into arthritis, in others, the excision seems to be more successful. However, the reason why some wrists develop postoperative instability while others do not remains unclear.[8] When the distal pole of the scaphoid is excised in the setting of a scaphoid nonunion, midcarpal instability may have already occurred so that this may be a unique situation compared with other indications for the procedure.

The structure of the midcarpal joint affects the transfer of forces through the wrist.[9] [10] Recent studies have identified two distinct structural patterns of the midcarpal joint. Type 1 wrists have a lunate that has a facet with the capitate alone, while the lunate in a type 2 wrist has a facet with both the hamate and the capitate.[11] Since the structure of the midcarpal joint affects the transfer of forces through the wrist,[9] [10] we hypothesize that the way forces are transferred through the wrist will be further altered when a distal or proximal pole excision is performed. Understanding these structural and mechanical patterns may help in identifying those wrists that will be prone to develop significant midcarpal joint instability, allowing a more individualized approach to treating scaphoid nonunion and wrist arthritis.

Materials and Methods

Sample Selection and Classification

Wrists were selected out of a database of normal wrist images that had corresponding normal computer tomography (CT) scans. Of the initial 58 sample CT images that were taken, 49 were of sufficient quality and were converted into .stl files to be imported into Fusion360. Due to the quality of the meshes from the CT scan, 19 patient samples were deemed feasible for simulation. These 19 CT scans that were selected from our patient database were anonymized and then converted into three-dimensional (3D) files. Nine samples were categorized as type 1 wrists, with the remaining 10 samples as type 2 wrists based on the bone morphology of the hamate, capitate, and lunate.[12] Wrist type was defined based on the lunate type in the midcarpal joint.[6]

Simulation Preparation

For the selected samples between type 1 (n = 9) and type 2 (n = 10) wrists, their respective mesh files were imported into the Fusion360 Design environment (Autodesk, v.2.0.14109). Initial sample mesh sizes were reduced and remeshed from an adaptive mesh to a uniform mesh. Face groups were generated to isolate each bone for repair before repairing them to create solid bodies. The bones were given the material properties of cortical bone based on the results from several literature sources.[12] [13] [14] Cancellous bone was omitted in this study, as the literature has shown minimal influence of cancellous bone on force transference and displacement in carpal bones.[13] The scaphoid was then split into proximal and distal halves by cutting the bone at a 40-degree angle from its base with the ulna.

Once the base model design was finished, it was brought into an implicit simulation workspace. A 200 N force was evenly split and applied to the dorsal crests of the trapezoid and capitate (100 N along each crest) to replicate the scenario of a person performing a push-up on their knuckles ([Fig. 1]). Our model design featured six degrees of freedom: three translational axes and three rotational axes. Structural constraints along the X- (coronal plane), Y- (sagittal plane), and Z-axis (axial plane) were applied to the bottom of the ulna and radius to prevent unwanted translational and rotational movement due to the applied force, providing a stable base for the carpus. The radius and ulna were also fixed in the three rotational axes, again to provide a stable base for the carpus. 3D placement of the bones was altered as little as possible, done only to prevent overlapping or move closer for the purposes of generating contact—the latter being done primarily for the pisiform.

Fig. 1 Application of 200 N load force spread across the crests of the trapezoid and capitate following excision of the distal pole.

Replicating the ligaments in the wrist model can be misleading due to variability between individual wrists and a paucity of data on ligament properties and anatomy. Furthermore, including viscoelastic material properties can be challenging and overly complicate the model. In this model, bonded contacts were used to replicate some of the stability that would be present from intact ligaments in the wrist, rather than relying on structural constraints at their anatomical anchor points. Thereby, we opted for an intact wrist model, assuming the ligaments and other connective tissues are uninjured. In unpublished data, we have validated this model against a separate model that does incorporate ligamentous structures. The contact detection tolerance (CDT) was adjusted for each model for the lowest tolerance that could be used for all three conditions: no scaphoidectomy and distal pole removal. Sample variability in CDTs was due to differences in bone morphology and position. The generated simulation mesh for each model was 10% of the original size of the initial model mesh. There were no viscoelastic materials present in the models.

Displacement Registration

After the completion of each simulation, the maximum displacement of the trapezoid, trapezium, scaphoid, capitate, and hamate was measured along each axis after the applied load. The anatomical directions for each axis were assigned as follows: the X-axis is radioulnar (positive values are radial, negative values are ulnar), Y-axis is anteroposterior (positive values are ventral motion, negative values are dorsal), and Z-axis is vertical (positive values are distal, negative values are proximal). The carpal bones of each model were constrained using CDT to determine when collision occurs and thereby determining the displacement. This modeling software has no way to optimize the CDT, and it had to be manually chosen for each model via trial and error. It is possible that tuning the models in this way could yield underconstrained models, leading to greater than anticipated displacements.

Due to widely varying displacement results, directly measuring the difference in carpal bone displacement did not return statistically significant values. The method used to determine the CDTs was not able to provide sufficient constraints for the solver to fully determine the system, and the attempt to return a numerical answer to an underdetermined system is what we believe to be the source of the overlarge displacements. Therefore, instead of raw displacements, we looked at the difference between simulation scenarios between and within models. To correct for the variations in order of magnitude, the displacements were normalized and reported as a percent change from an intact wrist.

Statistical Analysis

The simulation model was used to predict motion at the capitate and STT joint with excision of proximal and distal poles. Average movement values for each plan were calculated for each carpal bone stratified by type. An independent sample t-test was used to compare the average movement between type 1 and type 2 wrists. A paired-sample t-test was then calculated to assess average movement between baseline and proximal and baseline and dorsal. Percent change was then calculated for movement between type 1 and 2 wrists. All analyses were two-tailed and set at the p < 0.05 level. Analyses were performed using RStudio 6.1 for Windows.

Results

The average age in the type 1 wrist was 33.25 (standard deviation [SD] = 8.6) years and for the type 2 group 39.8 (SD = 8.8). Males were 60% in type 1 wrists and 80% in type 2 wrists.

The model predicted large differences in force transfer between a normal wrist and a wrist following distal scaphoidectomy (regardless of wrist type), as inferred by the displacement pattern of the carpal bones. This was 200% for all of the bones studied and in all directions.

Excision of the distal pole showed greater displacement in type 1 wrists for all carpal bones examined in our model. In type 1 wrists the change in motion of the trapezium trapezoid and capitate was larger as a whole than in type 2 wrists (p < 0.05). [Figs. 2] and [3] demonstrate type 1 and 2 wrist models before and after distal pole excision. These figures demonstrate the difference in force transfer between the two wrist types in a wrist with an intact scaphoid and a wrist following scaphoid distal pole excision ([Figs. 2] and [3]).

Fig. 2 Type 1 wrist displaying induced displacement from applied 200N load before and after (A, B) distal scaphoid pole removal. This is a type 1 wrist since there is no contact between the lunate and the hamate. The forces that the further study is needed to delineate the contribution of the various structures in any given wrist. Bones see are shown in color according to the scale at the side of the figure. The color pattern is different before and after scaphoid excision in the trapezium, trapezoid, capitate, and hamate.

Fig. 3 Type 2 wrist displaying induced displacement from applied 200 N load before and after (A, B) distal scaphoid pole removal. This is a type 2 wrist since there is significant contact between the lunate and the hamate. The forces that the bones see are shown in color according to the scale at the side of the figure. The color pattern is different before and after scaphoid excision in the trapezium, trapezoid, capitate, and hamate.

As stated, to correct for the disparities in order of magnitude, we looked at the percent change from baseline for each bone. Comparison between the percent difference in type 1 and type 2 wrist following scaphoid excision and baseline (intact wrist) demonstrated significant differences between trapezoid and trapezial motion (force transfer) ([Table 1]).

Table 1 Comparison of differences in displacement between type 1 and type 2 wrists

	Percent difference (SD)		p-Value
	Type 1	Type 2
Trapezium X	111.12 (79.55)	34.43 (59.26)	0.034
Trapezium Y	109.31 (80.90)	25.75 (36.77)	0.012
Trapezium Z	110.27 (81.59)	36.96 (67.64)	0.054
Trapezoid X	112.67 (81.18)	33.71 (60.75)	0.033
Trapezoid Z	111.49 (81.82)	17.62 (29.45)	0.005

Abbreviation: SD, standard deviation.

Note: Percent difference was calculated between intact and distal scaphoidectomy and then compared between a type 1 and type 2 model.

When considering the direction of motion (in the coronal, sagittal, or axial planes), we observed differences in the motion pattern of the overall wrist between both type 1 and type 2 and between distal excisions ([Table 2]). Upon excision of the distal pole of the scaphoid, the carpal bones in a type 2 wrist move in the same direction, (as a unit) ulnarly, dorsally, and proximally. In a type 1 wrist, the bones displace in different directions. Specifically, the trapezoid, trapezium, and hamate moved distally (the forces were directed distally) while the capitate moved volarly and radially. The trapezium moved radially. The scaphoid, however, moved as in a type 2 wrist, ulnarly, dorsally, and proximally ([Table 2]).

Table 2 Direction of change as predicted by the model

	Displacement		Direction
	Type 1	Type 2	Type 1	Type 2
Trapezoid
X	–2.22E + 13 ↓	–3.26E + 10 ↓	Ulnar	Ulnar
Y	–8.91E + 12 ↓	–1.91E + 10 ↓	Dorsal	Dorsal
Z	6.60E + 13 ↑	–8.28E + 10 ↓	Distal	Proximal
Trapezium
X	1.02E + 13 ↑	–5.66E + 10 ↓	Radial	Ulnar
Y	–9.12E + 12 ↓	–2.54E + 10 ↓	Dorsal	Dorsal
Z	6.76E + 13 ↑	–9.63E + 10 ↓	Distal	Proximal
Scaphoid
X	–3.62E + 13 ↓	–5.45E + 10 ↓	Ulnar	Ulnar
Y	–1.43E + 13 ↓	–1.90E + 10 ↓	Dorsal	Dorsal
Z	–7.45E + 12 ↓	–7.62E + 10 ↓	Proximal	Proximal
Capitate
X	–2.61E + 13 ↓	–5.87E + 10 ↓	Radial	Ulnar
Y	8.50E + 12 ↑	–2.46E + 10 ↓	Volar	Dorsal
Z	1.01E + 14 ↓	–5.28E + 10 ↓	Proximal	Proximal
Hamate
X	–3.29E + 13 ↓	–2.27E + 10 ↓	Ulnar	Ulnar
Y	–6.03E + 12 ↓	–1.52E + 10 ↓	Dorsal	Dorsal
Z	1.53E + 13 ↑	–1.52E + 10 ↓	Distal	Proximal

Note: The green arrows indicate a positive displacement from baseline, whereas red arrows represent a decrease.

Discussion

This study found that excision of the distal pole of the scaphoid affected the transfer of forces significantly in all wrists. We also found differences in force transfer between type 1 and type 2 wrists, specifically in the amount of force transferred (type 1 > type 2), in the percent difference from an intact wrist (type 1 > type 2) and in the direction of movement (type 1 bones move in different directions). These results suggest that a type 1 wrist may be more prone to develop midcarpal joint instability and arthritis, as the forces following distal pole excision are higher in this anatomical pattern and the midcarpal joint seems to be less stable (with the bones seeing more forces in different directions). Anatomically, a type 2 wrist is considered to have a more stable midcarpal joint due to the additional facet with the hamate.[12] [13] Rhee et al concluded that in proven scapholunate dissociation, type 2 lunates were associated with a lower incidence of dorsal intercalated segment instability.[15] [16] This study's finding that in a type 2 wrist the carpal bones move in the same direction with excision of the distal pole of the scaphoid, suggests that the bones move as one unit, held together by the more anatomically stable midcarpal joint. Further study is needed to establish the clinical association between wrist type and collapse following excision of the distal pole as well as to examine the transfer of forces more proximally to the proximal carpal row and radiocarpal joint when the distal pole of the scaphoid is excised.

A significant limitation of this study is that the model did not consider the mechanical contribution and importance of the ligamentous structures. This model is based on the assumption that since these are “normal” wrists, the ligamentous structures are intact and therefore the translation of the bones that is predicted by the model reflects the way the bones move/or the force that the bone “sees” upon load with intact ligaments sharing in the mechanical force transfer.

The other way to approach this issue is to model the ligamentous structures. The disadvantage of the latter method is that there is likely so much variability in the insertions of the ligaments, their thickness and length and physical properties (and we have very little experimental information regarding these), that we are making many assumptions not based on experimental data. It could be that these assumptions introduce more error than just assuming that the ligaments, whatever their properties in this individual wrist, are intact. At this time, both types of models have been validated against each other and this has been successful (unpublished data). That is not to say that either method is perfect, and more validation is needed. Specifically in cadaver studies, though the ligamentous properties of a cadaver may not correctly reflect the ligament properties in vivo.

In light of this, our results, demonstrating that the type 1 wrist moves differently than a type 2 wrist, may reflect differences in bony structure or differences in ligamentous structures or both to varying degrees. Distal pole excision involves disrupting the dorsal intercarpal ligament and the dorsal radiocarpal ligament. It is possible that this is the cause of midcarpal instability and not (or not only) the bony structure. Since there are no intracarpal ligaments between the capitate and the lunate, it is possible that disruption of the dorsal ligaments during distal pole excision would preferentially destabilize a less (bony) stable midcarpal joint, namely, a type 1 wrist. Further study is needed to delineate the contribution of the various structures in any given wrist.

The amount of bone removed during distal pole excision likely varies between individual cases. For example, excision of the distal fragment in a scaphoid nonunion may include a large part of the scaphoid, while arthroscopic excision for STT joint arthritis may include just the subchondral bone within the joint. It is possible that the amount of excision affects the degree of instability due to both bony and ligamentous destabilization. We suggest this is another question for further study.

There is a difference between the two groups regarding age and gender. This is not statistically significant due to the small sample size. Furthermore, there is no established mechanical difference between genders or age (in adult wrists) in the wrist though one study found differences in mechanical properties of the soft tissues surrounding the joints by age though in this study there was a mechanical difference in old age.[17] Further study is needed to better delineate the mechanical effect of age and gender on the mechanics of the joint.

In summary, this study submits that due to the unique biomechanical profile of each wrist type, midcarpal joint structure may be able to predict wrist behavior following excision of the distal pole of the scaphoid. This may allow the surgeon to refine the indications for this procedure. Further study is needed to translate the results into clinical practice.

In general, it may be important to be aware of wrist type, as the differing biomechanics could lead to different risk profiles. It should be noted that individual variance remains great. Moving toward a future of precision medicine, details such as these will be important in crafting personalized treatment plans that are optimized for each patient's unique biology.

Conflict of Interest

None declared.

• References

• 1 Garcia-Elias M, Lluch A. Partial excision of scaphoid: is it ever indicated?. Hand Clin 2001; 17 (04) 687-695x
• 2 Corbin C, Warwick D. Midcarpal instability after excision arthroplasty for scapho-trapezial-trapezoid (STT) arthritis. J Hand Surg Eur Vol 2009; 34 (04) 537-538
• 3 Malerich MM, Catalano III LW, Weidner ZD, Vance MC, Eden CM, Eaton RG. Distal scaphoid resection for degenerative arthritis secondary to scaphoid nonunion: a 20-year experience. J Hand Surg Am 2014; 39 (09) 1669-1676
• 4 Malerich MM, Clifford J, Eaton B, Eaton R, Littler JW. Distal scaphoid resection arthroplasty for the treatment of degenerative arthritis secondary to scaphoid nonunion. J Hand Surg Am 1999; 24 (06) 1196-1205
• 5 McCombe D, Ireland DC, McNab I. Distal scaphoid excision after radioscaphoid arthrodesis. J Hand Surg Am 2001; 26 (05) 877-882
• 6 Viegas SF. Limited arthrodesis for scaphoid nonunion. J Hand Surg Am 1994; 19 (01) 127-133
• 7 Garcia-Elias M, Lluch A, Saffar P. Distal scaphoid excision in scaphoid-trapezium-trapezoid arthritis. Tech Hand Up Extrem Surg 1999; 3 (03) 169-173
• 8 Catalano III LW, Ryan DJ, Barron OA, Glickel SZ. Surgical management of scaphotrapeziotrapezoid arthritis. J Am Acad Orthop Surg 2020; 28 (06) 221-228
• 9 Pendola M, Petchprapa C, Wollstein R. A preliminary model of the wrist midcarpal joint. J Wrist Surg 2021; 10 (06) 523-527
• 10 Wollstein R, Kramer A, Friedlander S, Werner F. Midcarpal structure effect on force distribution through the radiocarpal joint. J Wrist Surg 2019; 8 (06) 477-481
• 11 Viegas SF. The lunatohamate articulation of the midcarpal joint. Arthroscopy 1990; 6 (01) 5-10
• 12 Viegas SF, Patterson RM, Hokanson JA, Davis J. Wrist anatomy: incidence, distribution, and correlation of anatomic variations, tears, and arthrosis. J Hand Surg Am 1993; 18 (03) 463-475
• 13 Marques R, Melchor J, Sanchez-Montesinos I, Roda O, Rus G, Hernandez-Cortes P. Biomechanical finite element method model of the proximal carpal row and experimental validation. Front Physiol 2021; 12: 749372
• 14 Spatz HC, O'Leary EJ, Vincent JF. Young's moduli and shear moduli in cortical bone. Proc Biol Sci 1996; 263 (1368): 287-294
• 15 Rhee PC, Moran SL, Shin AY. Association between lunate morphology and carpal collapse in cases of scapholunate dissociation. J Hand Surg Am 2009; 34 (09) 1633-1639
• 16 Rhee PC, Moran SL. The effect of lunate morphology in carpal disorders: review of the literature. Curr Rheumatol Rev 2020; 16 (03) 184-188
• 17 Nguyen AP, Herman B, Mahaudens P, Everard G, Libert T, Detrembleur C. Effect of age and body size on the wrist's viscoelasticity in healthy participants from 3 to 90 years old and reliability assessment. Front Sports Act Living 2020; 2: 23

Publikationsverlauf

Eingereicht: 02. Januar 2024

Angenommen: 22. Februar 2024

Artikel online veröffentlicht:
14. März 2024