lunate advanced collapse (SLAC) of the wrist, and that the frequency of dorsal intercalated instability (DISI) increases with the longer duration of nonunion.
The relationship between the site of nonunion of the scaphoid and scaphoid nonunion advanced collapse (SNAC) H. Moritomo, K. Tada, T. Yoshida, T. Masatomi From the Kansai Rosai and Osaka Koseinenkin Hospitals, Japan
e studied retrospectively the radiographs of 33 patients with late symptoms after scaphoid nonunion in an attempt to relate the incidence of scaphoid nonunion advanced collapse (SNAC) to the level of the original fracture. We found differing patterns for nonunion at the proximal, middle and distal thirds. The mean intervals between fracture and complaint were 20.9, 6.7 and 12.6 years and obvious degenerative changes occurred in 85.7%, 40.0% and 33.3%, for the six proximal-, eight middle- and two distal-third nonunions, respectively. Nonunion at the proximal and middle thirds showed the first degenerative changes at the radioscaphoid joint, and this was followed by narrowing of the scaphocapitate and then the lunocapitate joints. In our two nonunions of the distal third degenerative changes were seen only at the lunocapitate joint. Most patients with SNAC and nonunion of the middle or distal third showed dorsal intercalated instability; few patients with nonunion of the proximal third developed this deformity. We discuss the initial management of nonunion of the scaphoid at different levels in the light of our findings, and make recommendations.
W
J Bone Joint Surg [Br] 1999;81-B:871-6. Received 27 July 1998; Accepted after revision 18 December 1998
Symptomatic nonunion of the scaphoid may lead to a type of degenerative arthritis with severe collapse which was termed ‘scaphoid nonunion advanced collapse’ (SNAC) by 1 Krakauer, Bishop and Cooney. The natural history of
H. Moritomo, MD, Orthopaedic Surgeon K. Tada, MD, Chief T. Yoshida, MD, Orthopaedic Surgeon Department of Orthopaedic Surgery, Kansai Rosai Hospital, 3-1-69 Inabaso, Amagasaki-shi, Hyogo 660-8511, Japan. T. Masatomi, MD, Orthopaedic Surgeon Department of Orthopaedic Surgery, Osaka Koseinennkin Hospital, 4-2-78 Fukushima, Fukushima-ku, Osaka-shi, Osaka 553-0003, Japan. Correspondence should be sent to Dr H. Moritomo at 1-12-5, ShinsenriKitamachi, Toyonaka-shi, Osaka 565-0081, Japan. ©1999 British Editorial Society of Bone and Joint Surgery 0301-620X/99/59333 $2.00 VOL. 81-B, NO. 5, SEPTEMBER 1999
2-5
nonunion of the scaphoid has been well reported, showing that the pattern of SNAC is similar to that of scapholunate advanced collapse (SLAC) of the wrist, and that the frequency of dorsal intercalated instability (DISI) increases with the longer duration of nonunion. There are few reports on the pattern of SNAC in relation to the site of the fracture. We have tried to identify the natural course of SNAC for various sites.
Patients and Methods We undertook a retrospective and radiological study of 33 symptomatic wrists with nonunion of the scaphoid seen between 1981 and 1997 at Kansai Rosai Hospital or Osaka Koseinenkin Hospital. No patient had been treated for the fractured scaphoid since the time of injury at a mean of 10.7 years (1 to 50) earlier. The mean age of the patients at review was 34.5 years (13 to 70); there were 28 men and five women. For each patient, the site of the fracture was determined 6 as described by Russe. Six nonunions were in the distal third, 20 in the middle third and seven in the proximal third of the scaphoid. Anteroposterior, lateral and oblique radiographs were taken and, when possible, tomography and CT were carried out. These were reviewed to locate the degenerative changes, record collapse of the proximal fragment and assess dorsiflexion of the lunate to determine carpal instability. The radiolunate angle on the lateral radiograph was measured with respect to the longitudinal axis of the radius. Its normal value is 7° of palmar flexion when the wrist is in a neutral position (12° of palmar flexion to 9° 7 of dorsiflexion). We recorded 10° or more of lunate dorsiflexion as abnormal. The location of degenerative arthritis was determined for 2-5 three areas: the radioscaphoid joint (area 1), the scaphocapitate joint (area 2), and the lunocapitate joint (area 3) (Fig. 1). The presence of degenerative changes was diagnosed from pointing of the radial styloid and by narrowing between the radius and distal scaphoid for area 1, between the capitate and the proximal scaphoid fragment for area 2, and between the capitate and the lunate for area 3. Progression of degenerative changes was related to the duration of nonunion, the frequency of DISI deformity, and the site of the original fracture. 871
872
H. MORITOMO,
K. TADA,
Fig. 1 Areas examined for radiological degenerative arthritis: 1) radioscaphoid joint 2) scaphocapitate joint, and 3) lunocapitate joint.
Results Our review of the radiographs of 33 patients with nonunion of the scaphoid showed that 16 patients had unequivocal degenerative changes (SNAC group) and 17 had none (nonSNAC group). The SNAC group included two of six distal-,
T. YOSHIDA,
T. MASATOMI
eight of 20 middle- and six of seven proximal-third nonunions. The relationships between the duration of nonunion, associated DISI deformity, and degenerative changes are shown in Table I. Most of the patients with nonunion of the proximal third already had degenerative changes when they first complained, while under half of those with nonunion of the distal or middle third had such changes. At the first examination, degenerative changes were more common after fractures of the proximal third (85.7%) than after injuries to the middle (40.0%) or distal thirds (33.3%). We also compared the non-SNAC group with the SNAC group. Table I shows that the mean age, duration of nonunion, and frequency of associated DISI deformity were all greater in the SNAC group, despite the finding that the nonSNAC group included more manual workers. In the SNAC patients, the progression of degenerative changes was related to the duration of nonunion and the frequency of DISI deformity (Table II). Each fracture level showed a different pattern of SNAC. In the fractures of the middle third, degenerative changes started between the distal fragment of the scaphoid and the radius (Fig. 1; area 1) as recognised by pointing of the radial styloid with or without dorsal radioscaphoid osteophytes (Fig. 2). Midcarpal arthritis was the next to appear (Fig. 3), shown first by narrowing of the joint between the capitate and the proximal fragment of the scaphoid (Fig. 1; area 2) and later (Fig. 4) between the capitate and the lunate (Fig. 1; area 3). The mean duration of nonunion in patients with radiographs showing degenerative changes in areas 1, 2 and
Table I. Summary of the 33 patients with late symptoms after nonunion of the scaphoid, with and without SNAC Non-SNAC patients
SNAC patients
All patients
Fracture site
Number of patients
Mean age (yr)
Mean interval (yr)
DISI (%)
Number of patients
Mean age (yr)
Mean interval (yr)
DISI (%)
Number of patients
Mean age (yr)
Mean interval (yr)
DISI (%)
Distal Middle Proximal Total
4 12 1 17
20.8 26.3 22.0 24.8
4.0 1.9 6.0 2.6
50 50 0 47
2 8 6 16
63.0 39.0 46.5 44.8
30.0 14.1 22.7 19.3
100 75 17 53
6 20 7 33
34.8 31.4 43.0 34.5
12.6 6.7 20.9 10.7
67 60 14 52
Table II. Findings in the 16 patients with SNAC
Case 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Fracture level Distal Middle
Proximal
Interval between injury and complaint (yr) DISI 24 36 3 4 6 14 18 20 18 30 10 10 9 22 35 50
+ + + + + + + + +
Sites of degenerative changes (see Fig. 1) Area 1
Area 2
Area 3
+ + + + + + + + + + + + + +
+ + + + + + +
+ + + + +
Mean interval (yr) 30 9
20 24 10 22 50
THE JOURNAL OF BONE AND JOINT SURGERY
THE RELATIONSHIP BETWEEN THE SITE OF NONUNION OF THE SCAPHOID AND SCAPHOID NONUNION ADVANCED COLLAPSE (SNAC)
873
Case 6. Radiographs of nonunion of a fracture of the middle third after 14 years showing a) changes in the radial styloid in the posteroanterior view (PA) and b) the lateral view with a DISI deformity.
Fig. 2a
Fig. 2b
Fig. 4a
Fig. 3
Figure 3 – Case 8. Radiograph of a nonunion of the middle third after 20 years showing pointing of the radial styloid in area 1 and narrowing in area 2. Figure 4 – Case 10. Radiograph of a nonunion of the middle third after 30 years showing a) a view with degenerative changes in areas 1, 2 and 3 and b) a lateral view with DISI deformity and degenerative change in area 3.
3 were 9, 20 and 24 years, respectively. Six of the eight patients with middle-third nonunion had DISI deformity; the other two had degenerative changes only in area 1. After fractures of the proximal third, the progression of degenerative changes was similar to that for the middlethird; the mean duration of nonunion in patients with changes in areas 1, 2 and 3 was 10, 22 and 50 years, respectively (Figs 5 to 7). Of the six patients with nonunion VOL. 81-B, NO. 5, SEPTEMBER 1999
Fig. 4b
of the proximal third, one had degenerative changes in all three areas, but the remaining five had no DISI deformity. After fractures of the distal third, the pattern and progress of the degenerative changes were completely different. The two patients with nonunion of the distal third developed degenerative changes only in area 3. After a mean duration of nonunion of 30 years, both had severe DISI deformity (Fig. 8).
874
H. MORITOMO,
K. TADA,
T. YOSHIDA,
T. MASATOMI
Case 11. Radiographs of a nonunion of the proximal third after ten years showing a) a PA view with pointing of the radial styloid in area 1 and b) a lateral view with no DISI deformity.
Fig. 5a
Fig. 5b
Case 14. Radiograph of a nonunion of the proximal third after 22 years showing a) a PA view with pointing of the radial styloid in area 1 and narrowing in area 2 and b) a lateral view with no DISI deformity.
Fig. 6a
Fig. 6b
Case 16. Radiograph of a nonunion of the proximal third after 50 years showing a) a PA view with narrowing in area 1 and collapse of the proximal fragment and b) a lateral view with DISI deformity.
Fig. 7a
Fig. 7b
Case 2. Figure 8a – A PA radiograph of a nonunion of the distal third after 30 years showing degenerative change in area 3 with no narrowing in areas 1 and 2. Figure 8b – Lateral tomograph showing a DISI deformity and confirming degenerative change in area 3.
Fig. 8a
Fig. 8b THE JOURNAL OF BONE AND JOINT SURGERY
THE RELATIONSHIP BETWEEN THE SITE OF NONUNION OF THE SCAPHOID AND SCAPHOID NONUNION ADVANCED COLLAPSE (SNAC)
Discussion The natural history of symptomatic nonunion of the scaphoid 2-5 is well documented. Degenerative changes after nonunion are similar to those in SLAC as described by Watson and 8 Ballet, but show some differences and are better termed 1 scaphoid nonunion advanced collapse or SNAC. Previous reports on SNAC stated that the frequency of DISI increased with a longer duration of nonunion and that 4 degenerative change may be early in the presence of DISI, but there is little information about the different patterns of SNAC in relation to the site of the fracture. Fractures of the middle third are much more common than in the distal and proximal thirds, and have therefore provided most of the information on SNAC. We found different patterns of SNAC for different fracture sites. For fractures of the middle third, we confirmed that DISI increases with time, and that degeneration starts between the radius and the distal scaphoid fragment. The scaphocapitate joint is affected later, followed by the capitolunate joint. After fractures of the distal third, DISI also increases with time, but degenerative change is seen only in the lunocapitate joint. Fractures of the proximal third rarely develop DISI, but degenerative changes progress in the same order as in the middle third. Only one of our six proximal nonunions developed DISI and this followed collapse of the proximal fragment. It seems probable that DISI and lunocapitate degeneration occur only when the proximal fragment of the scaphoid has collapsed. Our record of the duration of nonunion is from the time of the original injury to when the symptoms became severe enough for the patient to seek medical help. The high prevalence and degree of degenerative changes at this stage suggest that these types of nonunion may not produce symptoms until the radiological changes have become severe. The mean duration of nonunion, before complaint, was longer for fractures of the proximal and distal thirds than for those of the middle third. Degenerative changes were most common after fractures of the proximal third, suggesting that these produce less early symptoms than those of the middle third and that the symptoms start only after the changes have become severe. In our two nonunions of the distal third, both symptoms and degenerative changes were late to develop. In nonunion of the middle and distal thirds, more than half of the patients with complaints did not have obvious degenerative changes. Comparison between groups showed that the mean age, the duration of nonunion and the frequency of associated DISI were greater in the SNAC group than in those without SNAC. The latter included more heavy manual workers, indicating that activity may affect the development of symptoms. High activity in young manual workers may cause symptomatic synovitis before there is obvious degenerative arthritis. 8 In SLAC, rotation of the entire scaphoid results in loss of normal congruency between the elliptical articular surVOL. 81-B, NO. 5, SEPTEMBER 1999
875
face of the scaphoid and the scaphoid fossa of the radius, causing radioscaphoid arthritis. This develops first at the radial styloid, and subsequently along the entire scaphoid fossa. Separation between the scaphoid and the lunate allows proximal migration of the capitate, which pushes the radial side of the lunate proximally and laterally. Increased shear loading destroys the lunocapitate joint. In SNAC the pathomechanics are slightly different. In nonunion of the middle third, the distal fragment of the scaphoid has no ligamentous attachments. This allows it to rotate, causing displacement, angulation, or both, producing incongruity between this fragment and the articular surface of the radius. Initial degenerative changes are seen opposite the distal fragment, stopping at the site of nonunion, so that arthritis develops only around the fragment. The proximal fragment normally remains attached to the lunate by an intact interosseous ligament; it rotates with the lunate in their relatively spherical bed, allowing perpendicular loading of the articular cartilage in all positions. In nonunion of the distal third, rotational subluxation may occur as in the middle third, but this causes minimal incongruity because most of the fracture line is distal to the radioscaphoid joint. The proximal fragment is larger and most of the radioscaphoid surfaces remain congruent. Degenerative changes are probably rare and symptoms are late to develop. The most significant factors associated with degenerative arthritis were displacement of the fracture and instability, especially DISI deformity which increases with time after nonunion of the middle third. A DISI deformity concentrates the contact area and pressure in the lunocapitate joint on to the dorsal lip of the lunate. Degenerative changes at the lunocapitate joint progress slowly since there is no separation between the scaphoid and the lunate as seen in a SLAC wrist. In nonunions of the proximal third, the distal fragment of the scaphoid may maintain ligamentous attachments, especially the dorsal intercarpal ligament and the dorsal scapholunate interosseous ligament. These maintain the scaphoid in a neutral position and prevent the development of a DISI deformity. Such a nonunion, even without instability, may still develop degenerative changes if there is sufficient movement. The relative delay in onset of symptoms, even if degenerative changes exist, is because of the lack of the carpal instability seen after a nonunion of the middle third. The small number of patients which we studied makes it difficult to produce definitive conclusions, but some trends and patterns are clear. Our findings may help surgeons to advise treatment for early nonunion based on the different probability of subsequent degenerative changes at different levels of injury. It is important to restore anatomical position by correcting angulation and length for nonunion of the distal or middle third. We therefore recommend that DISI deformity should be reduced and grafted from the palmar side before
876
H. MORITOMO,
K. TADA,
degenerative changes have developed; chronic malunion will lead to late degenerative changes in the lunocapitate joint. For nonunion of the proximal third, bony union is more important than the correction of deformity, since this is rarely seen. Bone grafting or revascularisation from the dorsal side is recommended. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
References 1. Krakauer JD, Bishop AT, Cooney WP. Surgical treatment of scapholunate advanced collapse. J Hand Surg [Am] 1994;19:751-9. 2. Mack GR, Bosse MJ, Gelberman RH, Yu E. The natural history of scaphoid non-union. J Bone Joint Surg [Am] 1984;66-A:504-9.
T. YOSHIDA,
T. MASATOMI
3. Ruby LK, Stinson J, Belsky MR. The natural history of scaphoid non-union: a review of fifty-five cases. J Bone Joint Surg [Am] 1985; 67-A:428-32. 4. Vender MI, Watson HK, Wiener BD, Black DM. Degenerative change in symptomatic scaphoid nonunion. J Hand Surg [Am] 1987; 12-A:514-9. 5. Inoue G, Sakuma M. The natural history of scaphoid nonunion: radiographical and clinical analysis in 102 cases. Arch Orthop Trauma Surg 1996;115:1-4. 6. Russe O. Fracture of the carpal navicular: diagnosis, non-operative treatment, and operative treatment. J Bone Joint Surg [Am] 1960; 42-A:759-68. 7. Sarrafian SK, Melamed JL, Goshgarian GM. Study of wrist motion in flexion and extension. Clin Orthop 1977;126:153-9. 8. Watson HK, Ballet FL. The SLAC wrist: scapholunate advanced collapse pattern of degenerative arthritis. J Hand Surg [Am] 1984;9-A: 358-65.
THE JOURNAL OF BONE AND JOINT SURGERY
