Pencil Grip X-Ray View Improves Detection of Early Wrist Arthritis

Abstract

Objectives

Long-term sequelae of scapholunate instability predictably result in progressive arthritic changes. In early stages of arthritis, joint space narrowing, subchondral sclerosis, and osteophytic formation can be difficult to appreciate on posterior–anterior (PA) X-ray views of the wrist. In patients with scapholunate advanced collapse (SLAC) wrist, the pencil grip X-ray may more accurately define the joint space narrowing in the radioscaphoid and radiolunate articulations compared with standard PA radiographs.

Materials and Methods

We conducted a retrospective chart review of all patients with pencil grip X-ray views and documented wrist arthritis at a single institution. A total of 29 patients met criteria. Two authors independently reviewed wrist X-rays and classified the X-rays based on radiographic stages of arthritis for both PA and pencil grip X-rays. Radioscaphoid and radiolunate joint spaces were measured, and mean comparisons were performed using student's t -test, and boxplot comparisons were calculated for each arthritis grade.

Results

In direct comparison of joint space narrowing, the radioscaphoid distance observed on the pencil grip radiographs was significantly lower than the PA wrist view in all arthritis grades ( p  < 0.05). The pencil grip view demonstrated more accurate joint space narrowing in patients with grade 2 osteoarthritis, but, not in other grades. There were no differences in radiolunate distance between pencil grip and PA wrist X-rays in all arthritis grades.

Conclusion

The pencil grip X-ray view demonstrated significantly lower radioscaphoid distances across all arthritis grades compared with PA views, suggesting it may be more sensitive in detecting joint space narrowing. In particular, the pencil grip view more accurately demonstrated radioscaphoid joint space narrowing with grade 2 osteoarthritis. No significant differences were found in radiolunate distances between the two imaging methods.

Clinical Relevance

This study highlights the potential of the pencil grip X-ray view as a valuable tool for more accurately assessing early radioscaphoid wrist arthritis.

Long-term sequelae of scapholunate instability predictably result in progressive arthritic changes. Of the patients with symptomatic arthritis, surgical management is often tailored to the involved articular surfaces. In their review of 4,000 wrist X-rays in 1984, Watson and Ballet first described this progression of arthritis and likened the pathomechanics to two off-axis spoons. ¹ In the early stages of radiocarpal arthritis, evaluation of joint space narrowing and cartilage can be difficult. Several modalities have been suggested to supplement radiographs, such as magnetic resonance imaging (MRI), arthroscopy, and CT arthrography; however, there is no clear gold standard. ^{2 3}

The pencil grip X-ray view, also referred to as the clenched pencil view, involves the patient grasping a pencil with both hands, pronated 20 degrees and ulnarly deviated 30 degrees, with the forearms, wrists, and thumb metacarpals resting on the X-ray cassette. ^{4 5} This view has been shown to be an optimal, reproducible stress view for scapholunate dynamic instability. ⁴

The authors have anecdotally noticed that pencil grip X-ray views more accurately demonstrate radiocarpal joint space narrowing. In the early stages of arthritis, joint space narrowing, subchondral sclerosis, and osteophytic formation can be difficult to appreciate on posterior–anterior (PA) X-ray views of the wrist. In patients with scapholunate advanced collapse (SLAC) wrist, does the pencil grip X-ray more accurately define the joint space narrowing in the radioscaphoid and radiolunate articulations compared with standard PA radiographs? We hypothesized that the pencil grip X-ray view would more accurately detect joint space narrowing compared with standard PA views of the wrist.

Materials and Methods

We conducted a retrospective chart review of all patients with pencil grip X-ray views and documented wrist arthritis. Institutional Review Board approval for this study was obtained prior to initiation of chart review, and all necessary informed/HIPAA consent procedures were followed accordingly. A total of 547 patients had obtained pencil grip X-ray views, and of those, 29 patients had a documented history of wrist osteoarthritis. At our institution, pencil grip wrist X-rays are routinely obtained in new patients with wrist pain. Our technique for obtaining the pencil-grip view is demonstrated in Fig. 1 . Similar to how the technique was originally described, the patient holds a pencil and places bilateral wrists and thenar eminences on the cassette, which places the wrists in slight pronation and puts the scapholunate interval en face. All patients who were 18 years or older. We queried all patients with the following ICD-9 or ICD-10 codes for wrist osteoarthritis or wrist sprain (ICD-9: 715.13, 716.14, 718.84, 842.01; ICD-10: M19.031, M19.032, M19.039, M24.841, or M24.842) and individually reviewed each chart.

Fig. 1.

Clinical demonstration of the pencil grip X-ray technique.

Two authors independently reviewed wrist X-rays and classified the X-rays based on radiographic stages of arthritis for both PA and pencil grip X-rays. The true arthritis grade was based on the higher grade between the two scores. For each X-ray, the reviewer was blinded to the clinical outcome or joint space narrowing measurements. In addition, to determine the effectiveness of the pencil grip X-ray as an arthritis grading stress view, we calculated joint space narrowing using ImageJ with a previously described technique. ^{6 7} The middle of the radioscaphoid and radiolunate joint spaces were measured, and the middle of the capitate was measured as a control. The ratio was calculated for each joint space to account for magnification differences.

Interobserver reliability was calculated as 0.83, which is consistent with the literature. ⁸

Mean comparisons were performed using student's t -test, and boxplot comparisons were calculated for each arthritis grade. A p -value of less than 0.05 was considered significant. All calculations were performed using R Studio (2024; Boston, Massachusetts).

Results

In our cohort of 29 patients with wrist arthritis and pencil grip view, the average age was 53 ± 14.3 ( Table 1 ). The majority of patients had grade 2 arthritis ( n  = 11, 37.9%) and were male ( n  = 20, 69%). In direct comparison of joint space narrowing, the radioscaphoid distance observed on the pencil grip radiographs was significantly lower than the PA wrist view in all arthritis grades ( p  < 0.05; Table 2 ). The radiolunate distance was equivocal between the two radiographic views.

Table 1. Demographics.

	Mean ( n )	Std. dev. (%)
Age	53.3	14.3
Grade
0	5	17.2
1	7	24.1
2	11	37.9
3	5	17.2
4	1	3.4
Sex
Female	9	31
Male	20	69
Race
Black/African American	1	3.4
Other	2	6.9
Unavailable	2	6.9
White/Caucasian	24	82.8
Hand dominance
Left	1	3.4
Right	28	96.6

Table 2. Comparison of joint space narrowing as measured on both the pencil grip view and PA view.

	Mean ( n )	Std. dev. (%)	p -Value
Pencil grip radioscaphoid distance	4.5	4.2	0.004
PA radioscaphoid distance	6.1	3.3
Pencil grip radiolunate distance	9.2	2.4	0.8
PA radiolunate distance	9.3	2.8

Further stratification by arthritis grade is demonstrated in box plots with mean comparisons by grade for both radioscaphoid distance and radiolunate distance. The pencil grip X-ray view demonstrated more accurate joint space narrowing in patients with grade 2 osteoarthritis; however, not in other grades ( Fig. 2 ). There were no differences in radiolunate distance between pencil grip and PA wrist X-rays in all arthritis grades ( Fig. 3 ).

Fig. 2.

Boxplot graphs of radioscaphoid distance stratified by arthritis grade.

Fig. 3.

Boxplot graphs of radiolunate distance stratified by arthritis grade.

Discussion

Evaluation of joint space narrowing and cartilage in the carpus is a crucial aspect of diagnosing and treating symptomatic wrist arthritis. The size and complex anatomy of the wrist present unique challenges in assessing osteoarthritic changes reliably. MRI and arthroscopy have been proposed as more accurate evaluations of arthritis; however, few studies have focused on stress radiographs of wrist arthritis.

Similar to the Rosenberg flexed PA view of the knee, which has been used to demonstrate joint space narrowing in the medial compartment of the knee, the pencil grip X-ray view more accurately identifies joint space narrowing in the radioscaphoid articulation ( Fig. 4 ). ⁹ From the original description of the (clenched) pencil grip X-ray view, the hand is placed in 30 degrees of ulnar deviation and 20 degrees of pronation. ⁵ Ulnar deviation of the hand permits the scaphoid to extend, which leads to increased contact with the scaphoid fossa. ¹⁰ Pronation, which can facilitate more en face views of the radiocarpal joints, has minimal effect on scaphoid flexion-extension. ^{4 10}

Fig. 4.

Knee X-rays demonstrating the Rosenberg flexed PA view of the knee that is more accurate.

In his review of more than 4,000 wrist X-rays, Watson originally likened the SLAC pattern of arthritis to two superimposed malrotated spoons to describe the cartilage wear pattern and progression of arthritis. ¹ In his classification schema, one of the earliest indicators is the presence of joint space narrowing and subchondral sclerosis between the scaphoid and the radial styloid. These changes correspond to Watson's stage I SLAC wrist. As arthritis progresses to stage II, joint space narrowing and degenerative changes extend to involve the entire radioscaphoid joint.

Several studies have evaluated the reliability and validity of diagnosing arthritis on PA wrist X-rays, and in particular, poor reliability in diagnosing early stages of arthritis. Vishwanathan et al, in 2013, reviewed 41 radiographs to determine the reliability of wrist arthritis grading. ⁸ Four separate reviewers on two separate occasions graded each X-ray. Inter- and intraobserver reliability was graded as moderate (kappa coefficient = 0.65 and 0.59). However, the interobserver reliability improved as the SLAC grade increased (kappa coefficient = 0.47, 0.52, 0.69 for stages 1, 2, and 3, respectively), emphasizing the poor reliability of early arthritis grading. ⁸

Other modalities have been suggested to aid in diagnosing early cartilage wear in the wrist, such as MRI, CT arthrography, and diagnostic arthroscopy. In a study by Belhaouane et al, published in 2016, six graders (two orthopedic surgeons, two residents, and two radiologists) reviewed 50 wrists with CT arthrography and X-rays on two separate occasions. The authors demonstrated poor reliability and reproducibility between authors.

MRI is a valuable tool for evaluating cartilage in the wrist; however, its efficacy in detecting small cartilage defects in the carpal bones remains limited. The small size of the carpal bones and the thin layer of articular cartilage covering them make it challenging to visualize subtle injuries using standard MRI protocols. Studies have shown that conventional MRI techniques may not be sufficiently sensitive or accurate for diagnosing cartilage defects in the wrist and that detecting these defects can be provider dependent. ^{2 11 12 13 14 15}

Wrist arthroscopy has also been proposed as a superior method for evaluating cartilage injuries in the carpus. ^{14 16 17} The advantages of arthroscopic evaluation include direct visualization of the articular surfaces and the ability to directly cartilage to evaluate integrity directly, although arthroscopic evaluation is invasive and not routinely used prior to wrist arthritis surgery.

One of the main limitations of our study is the number of patients that fit our criteria of having a pencil grip X-ray view, as well as radiographic arthritis. In our review of 547 patients since our institution began obtaining the pencil grip X-ray view, only 29 cases had arthritis. This was likely due to our inclusion criteria, and there may be a component of selection bias. Older patients with presumed wrist arthritis were not routinely sent for pencil grip X-ray views. Though we believe this bias is mitigated by the finding that a pencil grip detects the earlier stages of arthritis rather than the later stages. Future studies should focus on a more comprehensive collection of arthritis stages as well as a larger number of patients.

The pencil grip X-ray view demonstrated significantly lower radioscaphoid distances across all arthritis grades compared with PA views, suggesting it may be more sensitive in detecting joint space narrowing. In particular, the pencil grip view more accurately demonstrated radioscaphoid joint space narrowing with grade 2. No significant differences were found in radiolunate distances between the two imaging methods. This study highlights the potential of the pencil grip X-ray view as a valuable tool for more accurately assessing early radioscaphoid wrist arthritis.