Revisiting the Natural History of Chronic Scaphoid Nonunions: A Retrospective Study of 20 Cases

Abstract

Background  The natural history of scaphoid nonunion is the development of degenerative arthritis. A lot of information is still unclear about this progression. The purpose of this study is to analyze patients with scaphoid nonunions who had not received any kind of treatment and to assess the functional outcome.

Materials and Methods  This is a retrospective study that analyzed the patients with chronic scaphoid nonunions between 2009 and 2019. None of the patients received any treatment. The age at the time of injury, examination, pattern of fracture, types of scaphoid nonunion, symptoms, and duration of nonunion were noted. Diagnosis was confirmed by radiographs, computed tomography (CT) scan, and magnetic resonance imaging (MRI). Scapholunate and radiolunate angles were recorded. Pain score, modified mayo wrist score, grip strength, range of movement, and the functional outcome of these scaphoid nonunions were analyzed. A statistical correlation between the scaphoid nonunion presentations and the functional outcome was assessed.

Results  The mean age of the patients was 62 years (range: 35–82 years.). There were 17 male and 3 female patients. There were 9 waist and 11 proximal pole scaphoid nonunions. The mean duration of scaphoid nonunion was 34 years (range: 10–62 years). None of the patients had avascular necrosis (AVN) of the proximal scaphoid. The age at examination, gender, side of injury, fracture pattern (waist/proximal pole), fracture displacement ≤ 1 mm or > 1 mm, nonunion duration, and radiographic arthritic parameters had no significant impact on the functional outcome.

Conclusions  Untreated chronic scaphoid nonunion leads to the development of degenerative arthritis over a period of years, which is still unpredictable. Most of the patients become aware of the nonunion following a precedent injury or other reasons. Most of the patients have fair/good functional outcome despite reduced range of movements and grip strength. Many do not favor surgical intervention in the course of nonunion. Chronic nonunions open a lot of unanswered questions.

Clinical Relevance  There have been numerous studies on the treatment aspects of scaphoid nonunion, with little knowledge about certain people with nonunion who did not have any kind of treatment. The demographics, clinical findings, and radiological parameters do confirm the progression of these nonunion to arthritis, but most of them had fair-to-good outcome throughout their life. It opens our thinking about the real need of treatment in such nonunions and raises numerous questions about the disease.

Level of Evidence  This is a Level IV study.

The incidence of scaphoid nonunion due to interruption of tenuous blood supply is 10 to 15%, and that in displaced scaphoid fractures is 55%. ^{1 2} There have been extensive studies on the treatment of scaphoid nonunion, which are still evolving among the differences, lack of consensus, and controversies. Degenerative arthritis, scaphoid nonunion advanced collapse (SNAC) wrist, and dorsal intercalary segmental instability (DISI) occur subsequent to chronic scaphoid nonunion and untreated cases. Less is known about the patient's functional outcome in these nonunions. Also, various aspects of patient's lifestyle change, pain, restriction of movements, desire not to get operated, and the pattern of scaphoid nonunion progression are not clearly understood.

The purpose of the study was to analyze the patient's factors, time duration of diagnosing nonunion, pattern of degenerative arthritis, and the outcome subsequent to scaphoid nonunion.

Materials and Methods

Patients, Setting, and Ethics

This retrospective study was approved by the Institutional Ethical Committee Review Board. A total of 20 consecutive patients with scaphoid nonunion were analyzed between 2010 and 2019. Informed consent was obtained from the patients for the pictures. Age at the time of injury and at examination, type of injury, pattern of fractures, displacement, treatment, if any, presence of pain, loss of motion and grip strength, functional status of the hand, and clinical examination were noted. Pain only with strenuous use of the wrist and rarely required medication was considered as “mild” symptom, pain that required rest and medication was considered as “moderate,” and pain at rest, requiring pain medication regularly, and avoiding manual work and sports requiring motion of the wrist were considered as “severe” symptoms. ³ Diagnosis was made with radiographs (posteroanterior, lateral, and oblique wrist views, and a posteroanterior ulnar deviation view), Computer tomography (CT) scans, and magnetic resonance imaging (MRI). Radiocarpal and midcarpal arthritis, proximal pole avascular necrosis (AVN), DISI deformity, joint space reduction between radius and scaphoid the displacement, were also noted. Lichtman's classification for Scaphoid nonunion was used in this study. ³ Associated carpal fractures, dislocations, and the presence of concomitant distal radius fractures were excluded from the study. The time interval between the injury and diagnosis was noted.

Statistics

Qualitative (gender and side) and quantitative variables (age) were evaluated using Student's t -test and Mann–Whitney/Wilcoxon two-sample test (Kruskal–Wallis test for two groups). Variables were compared using chi-square and Fischer's exact test. Two-tailed p  < 0.05 was considered significant.

Follow-Up

At follow-up, scapholunate angle (degree), carpal height index, and radiolunate angle were measured from the radiographs. Radiolunate angle (–10° to 12°), Scapholunate angle (30–60°), Carpal height ratio (0.51–0.57) were considered as normal values. Radiolunate angle > 15° was considered as DISI deformity. Radiocarpal arthritis and midcarpal arthritis were graded using the Kellgren-Lawrence system. ⁴ [grade 0 (none), grade ⁴ (severe)]. MRI with low-signal intensityon T1-weighted sequences, lack of contrast enhancement, and homogenously decreased signal on T2-weighted fat-suppressed images assessed the AVN of the proximal pole of the scaphoid. Also, active wrist range of motion and grip strength were measured using goniometer and Jamar Hydraulic Hand Dynamometer (Model J00105) (Sammons Preston, Bolingbrook, IL). Pain was assessed on a visual analog scale (VAS), and function was assessed using the Disability of the Arm, Shoulder, and Hand (Quick DASH) questionnaire (0: no limitation; 100: maximum limitation) and the Mayo wrist score ([MWS] 91–100: excellent; 80–90: good; 65–79: fair; and <65: poor).

Results

Patient and Nonunion Characteristics

The mean age of the patients was 62 years (range: 35–82 years.). There were 17 male and three female patients. Of the 20 patients, 13 had right wrist involvement. The mean age at the time of injury was 28 years (range: 17–38 years). The mechanism of injury was unknown in 16 patients. There were 9nine waist and 11 proximal pole scaphoid nonunions ( Fig. 1 ). The mean duration of scaphoid nonunion was 34 years (range: 10–62 years). Diagnosis was made when they presented to the hospital for occasional wrist pain ( n  = 8), weak grip ( n  = 4), pain and loss of motion compared with the normal wrist ( n  = 4), cellulitis ( n  = 2), and unrelated problem ( n  = 2). Mild symptoms were seen in 10 patients, moderate in 8, and severe in 2. Lichtman's classification of scaphoid nonunion stage II was seen in four patients and stages III and IV in eight patients each ( Table 1 ).

Fig. 1.

( A, B ) Posteroanterior and lateral radiographs of an 82-year-old man with left-sided scaphoid nonunion of 62 years' duration.

Table 1. Demographics.

S. no.	Side	Gender	Age at injury	Age at examination	Duration of nonunion (years)	Symptoms	Fracture pattern	Subjective acceptance for surgery
1	Right	Male	20	82	62	Mild	Proximal pole	No
2	Left	Male	22	75	53	Moderate	Proximal pole	Indecisive
3	Left	Female	30	65	35	Mild	Waist	No
4	Right	Male	35	78	53	Severe	Waist	Yes
5	Right	Male	25	69	44	Mild	Proximal pole	No
6	Right	Female	24	65	41	Moderate	Proximal pole	No
7	Right	Male	35	70	35	Moderate	Waist	No
8	Right	Male	25	55	30	Moderate	Proximal pole	Yes
9	Left	Male	35	68	35	Severe	Waist	No
10	Right	Male	27	45	18	Mild	Waist	No
11	Right	Male	22	50	23	Mild	Waist	No
12	Left	Male	30	75	45	Moderate	Proximal pole	Indecisive
13	Left	Male	22	39	17	Mild	Proximal pole	Indecisive
14	Right	Male	29	49	20	Mild	Waist	Indecisive
15	Left	Female	35	70	35	Moderate	Proximal pole	Yes
16	Right	Male	35	55	25	Moderate	Proximal pole	Indecisive
17	Left	Male	38	69	31	Moderate	Proximal pole	Indecisive
18	Right	Male	17	65	48	Mild	Proximal pole	No
19	Right	Male	24	50	26	Mild	Waist	No
20	Right	Male	25	35	10	Mild	Waist	No

Note: Symptoms: mild, pain only with strenuous use of the wrist and rarely required medication; moderate, pain that required rest and medication; severe, pain at rest, taking pain medication regularly, and avoiding manual work and sports requiring motion of the wrist. (Adapted from Mack et al. ³ )

The mean fracture displacement was 1.3 mm (range: 0.5–2 mm). The mean radiolunate angle was 13° (range: 0–30°), scapholunate angle was 76° (range: 60–90°), and carpal height ratio was 0.48 (range: 0.39–0.54) ( Tables 2–4 ). Of the 20 patients, 12 had DISI deformity. Radioscaphoid arthritis (80%), joint space reduction and narrowing or pointing of the radial styloid (80%), midcarpal arthritis (60%), bone resorptive changes and cyst resorption (70%), and carpal collapse (50%) were the common findings seen in the study. ( Table 3 ) None of the patients had AVN of the proximal pole of the scaphoid on MRI imaging ( Fig. 2 ).

Table 2. Radiographic parameters of instability.

S. no.	Type of nonunion	Fracture displacement mm (CT scan)	Radiolunate angle	Scapholunate angle	Carpal height ratio
1	III	1	0	60	0.54
2	IV	1.7	15	70	0.42
3	III	1.2	0	65	0.51
4	IV	1.8	25	90	0.41
5	III	1	10	65	0.53
6	IV	1.5	18	80	0.47
7	III	1.5	15	70	0.49
8	III	1	20	85	0.48
9	IV	1.5	30	90	0.42
10	III	1.5	5	60	0.53
11	IV	2	15	90	0.39
12	IV	2	24	90	0.41
13	II	0.5	15	80	0.47
14	II	0.75	15	90	0.45
15	IV	2	25	90	0.40
16	III	1	10	75	0.50
17	IV	1.5	15	80	0.46
18	III	0.75	5	60	0.54
19	II	1	0	60	0.54
20	II	1.5	0	70	0.54

Note: Classification of scaphoid nonunion: type I, simple nonunion (stable, nondisplaced, no degenerative change); type II, unstable nonunion (significant displacement or instability [dorsal intercalary segmental instability], no degenerative change); type III, nonunion with early degenerative change (radioscaphoid arthritis with joint space narrowing, subchondral sclerosis, pointing of radial styloid); type IV, scaphoid nonunion advanced collapse (arthritis in radio scaphoid and midcarpal joint); type V, scaphoid nonunion advance collapse plus generalized arthritis and radiolunate joint involved. (Adapted from Mack et al ⁸ .)

Table 3. Radiographic parameters of arthritis (Kellgren–Lawrence system grade).

S. no.	Radio scaphoid arthritis	Narrowing/pointing of radial styloid	Midcarpal arthritis	Carpal collapse	Bone resorption with cysts
1	Yes	Yes	Yes	No	Yes
2	Yes	Yes	Yes	Yes	Yes
3	Yes	Yes	No	No	No
4	Yes	Yes	Yes	Yes	Yes
5	Yes	Yes	No	No	No
6	Yes	Yes	Yes	Yes	Yes
7	Yes	Yes	No	Yes	No
8	Yes	Yes	No	Yes	No
9	Yes	Yes	Yes	Yes	Yes
10	Yes	Yes	Yes	No	Yes
11	Yes	Yes	Yes	Yes	Yes
12	Yes	Yes	Yes	Yes	Yes
13	No	No	No	No	Yes
14	No	No	No	No	Yes
15	Yes	Yes	Yes	Yes	No
16	Yes	Yes	Yes	No	Yes
17	Yes	Yes	Yes	Yes	Yes
18	Yes	Yes	Yes	No	Yes
19	No	No	No	No	Yes
20	No	No	No	No	No

Note: Kellgren–Lawrence classification system: grade 0 (none), definite absence of X-ray changes of osteoarthritis; grade 1(doubtful), doubtful joint space narrowing and possible osteophytic lipping; grade 2 (minimal), definite osteophytes and possible joint space narrowing; grade 3 (moderate), moderate multiple osteophytes, definite narrowing of joint space, and some sclerosis and possible deformity of bone ends; grade 4 (severe), large osteophytes, marked narrowing of joint space, severe sclerosis, and definite deformity of bone ends. (Adapted from Kellgren and Lawrence ⁴ .)

Table 4. Outcome.

S. no.	Wrist flexion	Wrist extension	Grip strength (% compared with normal side)	Wrist radial deviation	Wrist ulnar deviation	Mayo wrist score	VAS score	DASH score
1	70	40	90	15	30	80	1	4.2
2	60	30	70	10	15	75	3	8.3
3	75	45	92	20	20	80	0	4.2
4	60	20	75	10	10	60	5	17.5
5	65	30	88	20	30	75	1	6.7
6	70	30	86	20	35	80	0	5.8
7	75	30	90	20	35	70	2	13.3
8	75	35	86	20	20	70	1	5.8
9	50	15	50	10	10	60	5	12.5
10	60	40	92	20	30	80	0	8.3
11	65	40	90	15	30	80	0	4.2
12	60	25	85	15	25	75	2	6.7
13	75	40	92	20	30	80	0	5.8
14	70	35	90	20	35	70	1	4.2
15	60	20	80	15	20	80	1	8.3
16	70	35	90	20	30	80	0	4.2
17	50	20	70	15	20	60	2	10.8
18	70	45	86	20	30	80	0	4.2
19	75	45	94	20	40	80	0	4.2
20	75	45	95	20	35	80	0	4.2

Abbreviations: DASH, Disability of the Arm, Shoulder, and Hand; VAS, visual analog scale.

Fig. 2.

( A, B ) Magnetic resonance imaging (MRI) images showing the scaphoid proximal nonunion ( C, D, F ) with intercarpal arthritis and ( E ) absence of avascular necrosis (no low signal intensity on T1-weighted images).

Three of the 20 patients had a subjective feeling that they could have accepted surgery had they been diagnosed or visited the doctor earlier. Eleven preferred not to get operated despite the present conditions.

Outcomes

Of the 20 patients, 17 with various types of nonunion and different patterns of degenerative arthritis had fair-to-good outcome ( Fig. 3 ). The age at examination, gender, side of injury, fracture pattern (waist/proximal pole), fracture displacement ≤ 1 mm or > 1 mm, and nonunion duration had no significant impact on the functional outcome ( Table 5 ). Lichtman's types of nonunion (II, III and IV) and radiographic arthritic parameters (radioscaphoid arthritis, midcarpal arthritis, carpal height ratio, resorption/cyst changes, pointing of styloid) had no significant impact on the functional outcome. The age at the time of injury and radiolunate angle had a significant impact on the outcome. The radiographic arthritic parameters, angles, and range of motion had no significant impact on the outcome scores. The mean range of wrist flexion was 88%, extension was 50%, radial deviation was 68%, ulnar deviation was 60%, and grip strength was 85% of the opposite side. Median VAS pain score was 1.2 (range: 0–5). Median quick DASH score was 7.2 (range: 4.2–17.5) and Mayo wrist score was 75 (range: 60–80).

Fig. 3.

The range of movements in the left wrist. ( A ) Limited dorsiflexion, ( B ) normal palmar flexion with ( C ) limitation of ulnar deviation, ( D ) and radial deviation.

Table 5. Correlation of multiple variables with outcome.

Variables	Mean (SD/ t -value)	MMWS (fair/good)	Poor	p -Value
Age	62 (13.3)			0.15 a
< 65 y		11	0	0.03 b
> 65 y		6	3
Age at injury	27.8(6.0)	17	3	< 0.05 a
Side
Right		12	1	0.12 a
Left		5	2
Gender
Male		14	3	0.45 a
Female		3	0
Fracture pattern
Waist		7	2	0.44 d
Proximal pole		10	1
Type of nonunion
II		4	0	0.08 c
III		8	0
IV		5	3
Symptoms
Mild		10	0	< 0.05 c
Moderate		7	1
Severe		0	2
Fracture displacement
≤ 1 mm		8	0	0.25 c
> 1 mm		9	3
Time from injury to nonunion
≤ 35 y		11	2	1.0 d
> 35 y		6	1
Radiolunate angle
≤15°		13	1	<0.05 c
>15°		4	2
Scapholunate angle
≤ 75°		10	0	0.09 c
> 75°		7	3
Carpal height ratio
≤ 0.47		5	3	0.48 c
> 0.47		12	0
Radio scaphoid arthritis				0.35 c
Narrowing/pointing of radial styloid				0.35 c
Midcarpal arthritis				0.13 c
Bone resorption with cysts				0.23 c

^aT-test.

^b Chi-square statistics (two-tailed p -value).

^cMann–Whitney/Wilcoxon two-sample test (Kruskal–Wallis test for two groups).

^dFisher's exact analysis.

Discussion

Patients with chronic scaphoid nonunion are minimally symptomatic and tolerate their symptoms well. ⁵ The common presentations to the hand surgeon are pain, dorsal swelling, and unrelated problems. Clinical findings include anatomical snuffbox tenderness, dorsal swelling, loss of dorsiflexion, and decreased grip strength. ⁶ Patients present years to decades after fracture and the nonunion. In our study, patients with severe symptoms (pain at rest, taking pain medication regularly, and avoiding manual work and sports requiring motion of the wrist) had significantly poor outcome.

London observed that the severity of arthritis increases with time in scaphoid nonunions. ⁷ Many untreated symptomatic scaphoid nonunions undergo a specific sequence of degenerative changes, and in a sufficient time, development of arthritis occurs. ⁸ Widening of the fracture cleft, cyst formation, radio scaphoid arthritis, sclerosis of the fracture surfaces, and generalized arthritis of the wrist are hallmark radiographic signs of nonunion. ^{8 9} Most of the patients have well-established arthritis by 20 years and extensive arthritis by 30 years post-nonunion. ⁸ In our study, we noted 80 % of radio scaphoid arthritis, joint space reduction and narrowing or pointing of the radial styloid, 60% of midcarpal arthritis, 70% of bone resorptive changes and cyst resorption, and 50% of carpal collapse. We also noted that the nonunion duration (time interval between injury and the presentation) < 35 years and > 35 years had no significant impact on the outcome.

Fisk that noted instability is the cause of nonunion and early degenerative changes. ¹⁰ The most significant factors associated with arthritis are fracture displacement and wrist instability. The fracture displacement has a significant impact on the degenerative changes after 10 years. ⁸ In our study, we noted that fracture pattern and types of nonunion (Lichtman's II, III, and IV) produced degenerative arthritis invariably in all patients but had no significant impact on the outcome.

Ruby et al studied the natural history of scaphoid nonunion in 55 patients and reported AVN in 2 wrists and intercarpal collapse in 12 wrists. ¹¹ Proximal pole fractures are more likely to progress to nonunion because of decreased blood supply and has an incidence of 16 to 42% of developing AVN. ¹² Of the 20 patients, 11 had proximal pole nonunion in our study and none had AVN in the MRI imaging.

Undisplaced scaphoid fractures had DISI deformity and distal pole of scaphoid arthritis. Displaced fractures had scaphocapitate and scapholunate arthritis in addition to the distal pole of scaphoid arthritis. ¹¹ Association of ligamentous injuries with scaphoid fractures frequently causes DISI deformity and decreased vascularity contributing to scaphoid nonunion. ⁹ The DISI deformity occurs in 22 to 68% of scaphoid nonunions. ^{6 13} The lunate dorsiflexion depicts the displacement and carpal instability. Severe degenerative arthritis may mask the displacement and the DISI deformity. ⁸ In our study, DISI deformity was noted in 60% of the patients. Also, the scapholunate and radiolunate angles confirmed the DISI deformity and instability, and carpal height ratio confirmed the intercarpal collapse subsequent to instability and arthritis. Radiolunate angle > 15° had significantly poor outcome in our study.

The stable nonunions are firm fibrous nonunions, which prevent the deformity and show minimal cystic and arthritic changes in the radiographs. Long-standing stable nonunions become unstable over time resulting in pseudoarthrosis, carpal collapse, sclerosis, synovial erosions, fibrous cysts, and degenerative arthritis. ⁸ The fracture fragments undergo resorption during the nonunion, and the prognosis of such established scaphoid nonunions depends on the viability of the bone fragments and secondary changes. ^{13 14} In our study, we noted that scaphoid nonunion duration (<35 or >35 years) had no impact on the outcome. We also documented that they became unstable nonunions during the course of time and had resorption of the fracture fragments (carpal height ratio) and secondary arthritic changes. Also, the fracture displacements (>1 mm) and radiographic arthritic parameters had no significant impact on the outcome. All patients had viable bone fragments without necrosis. Of the 20 patients, 17 (85%) achieved fair-to-good outcome in this study.

Our study noted that chronic scaphoid nonunion developed degenerative arthritis in the radioscaphoid and midcarpal joints and led to secondary changes in the wrist. None of the patients had any kind of treatment, and 11/20 patients did not retrospectively prefer treatment. This raises a question: “Do all scaphoid nonunions need treatment?” We do not have an answer to this question, yet it raises an intriguing opportunity for the hand surgeon to introspect about scaphoid nonunion treatment protocols.

Current treatment options include internal fixation with bone grafting, vascularized bone grafting, midcarpal arthrodesis, four-corner fusion, and total wrist fusion. Surgical treatment for chronic nonunions poses potential risk for infection, ankylosis, decreased range of motion, persistent nonunion, and implant failure. ^{6 14} Proximal row carpectomy is more favorable for advance SNAC for patients who require less grip strength at work. Symptoms and functions can be improved by wrist denervation, radial styloidectomy, and osteophyte excision. ¹⁴ Because of severe pain, restriction of movements, and avoidance of manual work, three patients in our study retrospectively preferred the surgical options.

The limitations of our study are the small sample size and the retrospective design.

Our study revisited the natural history of chronic scaphoid nonunions and found that 85% of patients are capable of doing their daily activities with reduction in grip strength and wrist movements. They had occasional pain, and the outcome was fair and good. All patients are prone to develop instability, carpal collapse, and degenerative changes despite the age at the time of examination and nature of the symptoms. The longer duration of nonunion does not affect the outcome or their working ability. The increased age at the time of injury (>35 years) and increased radiolunate angle (>15°) were found to have significantly poor outcome in this study. We need a large prospective study and a control study to analyze the treatment modalities based on duration of nonunion, symptoms, types of nonunion, fracture displacement, and various aspects affecting the outcomes. The scaphoid nonunion of long duration introduces more questions than the answers:

• Does anyone with a long duration of scaphoid nonunion need treatment?

• Do scaphoid nonunion progression and degenerative arthritis have an end point?

• What are the clinical and radiological parameters to say that these patients with long-standing nonunion will benefit from surgery or nonoperative methods?

• Do we need to have a separate entity called “happy scaphoid nonunions” where patients did fairy well in their entire life with nonunions and did not prefer surgery?

• Does the classification, criteria of fracture displacements, radiolunate, scapholunate angles, carpal height ratio, arthritis findings, and various surgical modalities such as four-corner fusion and proximal row carpectomy become obsolete in these chronic nonunions?

• Is there any difference in the pathogenesis of AVN between acute scaphoid fractures and chronic scaphoid nonunion?

Funding Statement

Funding None.