- Email: [email protected]
Complications in Dorsal Percutaneous Cannulated Screw Fixation of Nondisplaced Scaphoid Waist Fractures
Complications in Dorsal Percutaneous Cannulated Screw Fixation of Nondisplaced Scaphoid Waist Fractures Brandon D. Bushnell, MD, Andrew D. McWilliams, MPH, Terry M. Messer, MD From the Department of Orthopaedic Surgery, University of North Carolina School of Medicine, Chapel Hill, NC; and WakeMed Health & Hospitals, Raleigh, NC.
Purpose: With advances in tools and techniques, percutaneous screw fixation of nondisplaced fractures of the scaphoid waist has gained increasing popularity in recent years as an alternative to prolonged cast immobilization or open reduction and internal fixation. Many reports cite low complication rates, including no complications in some series. The purpose of this study was to evaluate the complications encountered with dorsal percutaneous cannulated screw fixation of nondisplaced scaphoid waist fractures. Methods: A retrospective chart review was performed for 24 patients who had surgery performed by a single surgeon over a 5-year period. All cases involved dorsal percutaneous cannulated screw fixation of nondisplaced (⬍1 mm) fractures of the scaphoid waist. Complications were rated a priori as major or minor based on modifications of established criteria. Results: The overall complication rate was 29%; there were 21% (5/24) major complications and 8% (2/24) minor complications. Major complications consisted of 1 case of nonunion, 3 cases involving hardware problems, and 1 case of postoperative fracture of the proximal pole of the scaphoid. Minor complications included intraoperative equipment breakage— 1 case involving a screw and 1 case involving a guide wire. Conclusions: Complications in dorsal percutaneous cannulated screw fixation of scaphoid fractures may be more common than previously reported. (J Hand Surg 2007;32A:827– 833. Copyright © 2007 by the American Society for Surgery of the Hand.) Type of study/level of evidence: Therapeutic IV. Key words: Scaphoid, complications, scaphoid fractures, percutaneous screw fixation, percutaneous technique, dorsal percutaneous scaphoid technique.
istorically, nondisplaced fractures of the scaphoid waist were often managed with immobilization as the standard of treatment. Recently, there has been an increased trend toward surgical fixation of nondisplaced fractures of the scaphoid with percutaneous techniques. Several reports of successful results—most of them using a volar approach— have appeared in the literature. Many of these series have reported low complication rates (Table 1).1–18 In particular, there have been 4 reports6,7,15,18 involving the dorsal approach, with reported ranges of complications from 0% to 14%. Advances in tools and techniques, coupled with a desire to eliminate the long periods of immobilization associated with nonsurgical treatment, have con-
H
tributed to the popularity of the percutaneous technique. Some surgeons have even suggested that percutaneous treatment should be considered the new standard of care for this injury.19 The purpose of this study was to evaluate and report the complications encountered with dorsal percutaneous cannulated screw fixation of nondisplaced scaphoid waist fractures.
Materials and Methods After obtaining appropriate institutional research review board approval, a retrospective review of medical records was performed for all patients who had surgical fixation of any scaphoid fracture by the senior hand surgeon at our institution (T.M.M.), an The Journal of Hand Surgery
827
828
The Journal of Hand Surgery / Vol. 32A No. 6 July–August 2007
Table 1. Selected Reports on Percutaneous Fixation of Scaphoid Fractures Study
Year
N
Union Rate, % (n)
Woazasek and Moser1 Haddad and Goddard2 Adolfsson et al3
1991 1998
146 15
89 (130) 100 (15)
Volar Volar
2001 2001 2001 2001 2002 2002 2003 2005 2005 2005 2005 2005 2005
23 11 32 16 27 49 13 11 44 26 51 15 7
96 100 100 100 100 100 100 100 100 96 98 100 100
Volar Volar Volar Dorsal Dorsal Volar Volar Volar Volar Volar Hybrid Volar Dorsal
2005
52
96 (50)
Volar
8 (4)
2006 2007
21 18
95 (20) 94 (17)
Volar Dorsal
19 (4) 6 (1)
—
577
96%
—
8%
Bond et al4 Saeden et al5 Slade et al6 Slade et al7 Yip et al8 Jeon et al9 Chen et al10 Dias et al11 Drac et al12 Pirela-Cruz et al13 Shih et al14 Slade et al15 Wong et al16
Approach
(22) (11) (32) (16) (27) (49) (13) (11) (44) (25) (50) (15) (7)
Complication Rate, % (n) 9 (13) 7 (1) 8 9 6 0 0 0 8 9 30 4 2 0 14
(2) (1) (2) (0) (0) (0)* (1) (1) (13) (1) (1) (0) (1)
Arora et al17 Bedi et al18 Totals/average
Complications See below Radial Nerve Dysesthesia Nonunion, reflex sympathetic dystrophy Prominent Hardware See below None None None* RSD Superficial skin infection See below Nonunion Nonunion None EPL Rupture⫹ Nonunion (2), Delayed Union (2) Nonunion, superficial wound infection, CRPS (2) Nonunion Nonunion most common complication
The numbers included represent (when specified by the researchers) acute fractures with full follow-up evaluation: Woazasek and Moser1: nonunion (5), superficial radial nerve irritation (3), superficial wound infection (2), Sudeck’s atrophy (2), loose/painful screw (1). Saeden et al5: loose screw (1), failure of the screw to enter the proximal fragment (1). Dias et al11: hypertrophic scar (4), sensitive and hypertrophic scar (3), sensitive scar (3), superficial wound infection (1), hypoesthesia in region of palmar cutaneous branch of median nerve (1), mild early algodystrophy (1). † The extensor pollicis longus rupture was due to a dorsal plate used to treat a concomitant ipsilateral distal radius fracture. *These researchers did not consider hardware removal a postoperative complication.
academic-affiliated level I trauma center. Inclusion criteria included a nondisplaced (⬍1 mm) fracture of the scaphoid waist, a time from injury of less than 10 weeks, treatment involving dorsal percutaneous screw fixation as described by Slade et al,6,7,15,20,21 and at least a follow-up period of at least 3 months. Exclusion criteria were initial fracture of the proximal or distal pole, fracture displacement of more than 1 mm, carpal instability, a delayed diagnosis of longer than 10 weeks, the presence of an established nonunion or malunion, use of open technique, and a follow-up period of less than 3 months. Concomitant injuries were not considered exclusion criteria. Within the 5-year period between October 1, 2001 and September 30, 2006, 24 scaphoid fractures in 24 patients met our criteria and were included in this study. The average follow-up interval for these patients was 9 months (range, 3–24 mo). The average age of these 24 patients at the time of surgery was 28 years (range, 14 –53 y). Fifteen surgeries were performed on the left hand and 9 on the right. The
dominant hand was involved in 12 (50%) of the patients. Twenty-three cases involved the Accutrak screw system (Acumed, Inc., Beaverton, OR), and 1 case involved the TwinFix screw system (Stryker Leibinger Inc., Kalamazoo, MI). The primary data points for this study were the incidence and type of complications. We classified both intraoperative and postoperative complications into major and minor groups, based on a modified version of the method of the assessment of complications in wrist arthroscopy described by Beredjiklian et al.22 Major complications included any complication resulting in the need for additional surgical intervention, nonunion, malunion, fracture, compartment syndrome, permanent nerve injury, septic arthritis, vascular injury, complex regional pain syndrome, permanent stiffness, or tendon rupture. Minor complications included any complication that required additional nonsurgical intervention, resulted in deviation from the standard surgical technique, prolonged the surgical time, led to superficial
Bushnell, McWilliams, and Messer / Scaphoid Complications
829
infection or prolonged wound drainage (⬎5 days), or caused transient neurologic symptoms, stiffness, or tendon irritation. Because our definition of complications included nonunions and malunions, radiographic healing was assessed and defined as the lack of a visible radiolucent fracture line on any one of a minimum of 4 views of the scaphoid, as judged by the authors and by the radiologist’s report.23 Computed tomography (CT) scans could not be obtained for every patient and therefore were not used as the primary criterion for judging healing, but CT images were used to confirm plain radiographic findings in each patient for whom they were available.
Results Major Complications There were 5 major complications (21%), all occurring in the postoperative period: 1 case of nonunion, 3 cases involving hardware problems, and 1 case of postoperative fracture of the proximal pole of the scaphoid. The case of nonunion was diagnosed in a 38-yearold man, a smoker, 8 months after surgery for the initial injury. It was treated with screw removal and
Figure 1. Errant screw placement. A sagittal CT image shows dorsal malpositioning of the screw with inadequate capture of the distal fragment.
Figure 2. Delayed union and settling of the screw. An anteroposterior radiograph of a 30-year-old male smoker whose fracture had not healed at 11 months after percutaneous fixation. He was taken to surgery for a suspected nonunion, but the fracture was found to be rigidly healed at surgery. As shown in this radiograph, however, the screw had settled during the delayed time to union and remained prominent at the proximal pole. Screw removal resulted in resolution of the patient’s symptoms.
revision open reduction, internal fixation, and bone grafting through a volar approach. The patient then went on to heal over the course of several additional months. Three cases involved problems with the screws, necessitating an additional operation. One patient’s postoperative CT scan showed errant screw placement with dorsal malpositioning and inadequate capture of the distal fragment (Fig. 1). This patient healed after treatment with hardware removal and revision percutaneous screw fixation via a volar approach. Another patient’s screw was too long distally and caused symptomatic irritation. A third patient had a delayed union that allowed the screw to settle and serve as a source of pain (Fig. 2). These 2 patients’ symptoms resolved with removal of the screws. A unique complication occurred in a patient in whom the fracture had radiographically and clini-
830
The Journal of Hand Surgery / Vol. 32A No. 6 July–August 2007
Figure 3. Development of a proximal pole fracture. (A) A 30-year-old man had percutaneous fixation with a 20-mm Accutrak screw, and the fracture healed at 8 weeks. (B) At 11 months after surgery, however, he still complained of pain and stiffness in the wrist. Radiographs showed a fracture of the proximal pole of the scaphoid. The exact cause of this fracture remains unclear.
cally healed at 8 weeks after percutaneous screw fixation (Fig. 3A); the patient enjoyed several months of asymptomatic, normal function. At approximately 10 months after surgery, he began to have gradually worsening pain and stiffness despite no additional trauma to the wrist. Radiographs 11 months after surgery showed a fracture of the proximal pole of the scaphoid (Fig. 3B). Arthroscopic surgery to evaluate the proximal scaphoid and hardware removal was offered, but the patient instead requested nonsurgical immobilization to attempt to address his symptoms. Minor Complications There were 2 minor complications (8%), both occurring during surgery and both rated as minor because they involved deviation from standard technique and prolonged the surgery time.22 In the 1 case involving the use of the TwinFix screw system (Stryker Leibinger Inc.), the screw broke during insertion via separation of the proximal screw head from the screw shaft. An extensive effort was needed to remove the partially inserted broken shaft, followed by conversion to a larger
screw size. Another case involved breakage of a guidewire during retrograde passage, resulting in approximately 15 minutes of prolonged surgical time for wire removal (Fig. 4).
Figure 4. Intraoperative breakage of the guidewire. Failure to maintain the wrist in flexion can lead to breakage of the guidewire. In this case, the wire impinged on the dorsal wrist capsule during readvancement of the wire back out of the dorsal entry wound. Slight flexion of the wrist then broke the wire. The small wire fragment required approximately 15 minutes of additional surgical time to remove.
Bushnell, McWilliams, and Messer / Scaphoid Complications
Discussion Percutaneous treatment of scaphoid fractures has long been investigated as an alternative to open techniques. Streli24 first described the use of the percutaneous technique in 1970. Since that time, many additional reports have described the use of percutaneous fixation of scaphoid fractures, including several recent series1–18 with low complication rates (Table 1). Our results show higher complication rates than most of these contemporary reports,1–18 especially those using the dorsal technique. Complications reported for the dorsal technique range from 0% to 14%, yet we report a total rate of 29%.6,7,15,18 In 4 previous studies, 6,7,15,18 only 2 complications were reported out of 67 total patients, representing an overall rate of 3%. One reported complication consisted of an extensor pollicis longus tendon rupture secondary to irritation from a dorsal plate used to treat a concomitant ipsilateral distal radius fracture— essentially a problem not related at all to the treatment of the scaphoid fracture.15 The other reported complication consisted of a case of nonunion18—a problem also seen in our study and the most common complication overall in all reports of the percutaneous technique using any approach (Table 1).1–18 There are several potential reasons for the observed discrepancy in complication rates. One of the reasons is operator error. In the minor complication involving the broken guidewire, the wrist was not completely flexed to 90° during the retrograde passage of the wire, which backed out through the dorsal skin incision, resulting in wire impingement and breakage as it was advanced under power. The risk of complications such as this one can be minimized by meticulous adherence to technique. Another reason is equipment failure. In the case of the broken TwinFix screw (Stryker Leibinger Inc.), the components separated during the insertion of a screw that otherwise appeared normal. Such problems are likely unforeseeable and thus unpreventable. Some of the complications we observed may relate to the theory underlying the dorsal approach. With a dorsal approach, the radioscaphoid joint is violated by the drill, and any screw left prominent proximally can damage the joint surface of the scaphoid fossa of the radius, a key articular surface of the wrist. The newly described potential for delayed presentation of a proximal pole fracture of the scaphoid (Fig. 3) raises the issue of whether placing a drill into this known high-risk area of the bone is a good idea. Indeed, perhaps the fracture was due to unrecognized
831
avascular necrosis of the proximal pole caused by the drill. Furthermore, it is unknown what other longterm sequelae may result from making a large hole in the proximal scaphoid articular surface. Our case involving excessive distal screw length highlights another issue in the analysis of complications common to both the volar and distal techniques—the potential interaction of technique and equipment in measuring the appropriate, or perhaps inappropriate, length of the screw. We have found that the depth gauge used with the screw set routinely overestimates the required length for proper screw placement. A recent study by Bedi et al18 advocated subtracting 4 mm from the measured length. Similarly, we have found it necessary to subtract 6 mm from the given depth gauge length and in addition to confirm this number using a parallelguidewire technique (Fig. 5). The case of the proximal pole fracture highlights the danger of erring too far in the other direction and underestimating screw length (Fig. 3). In this case, it could be that the stress-riser created by the short screw led to the development of a fracture and worsened symptoms. Achieving an exact appropriate screw length within the bone—not too short, but not too long—thus appears to be critical to success and decreased postoperative complications. In cases involving tapered screw design, such as the Accutrak system (Acumed, Inc.) used in most of our patients, it becomes even more critical to measure screw length appropriately. If a screw’s excess
Figure 5. Use of an extra guidewire to confirm length. A second guidewire of equal length is placed directly parallel to the first wire to confirm the measured length for the screw. The portion of the second wire that extends beyond the first wire equals the interosseous portion of the first wire and thus should correlate with the measurement obtained from the depth gauge.
832
The Journal of Hand Surgery / Vol. 32A No. 6 July–August 2007
length necessitates replacement with a shorter screw, the initial interfragmentary compression and overall construct rigidity are considerably reduced because of the size mismatch between the second screw’s threads and the original path in the bone. Frequent fluoroscopic images during screw insertion can help the surgeon predict excess length, so that a smaller screw can be substituted if needed before completely seating the first screw. These potential issues with the dorsal technique raise the question of which percutaneous approach (dorsal/ proximal/antegrade or volar/distal/retrograde) is the best choice for minimizing complications. Complication rates ranging from 0% to 30% have been reported in series using a volar approach (Table 1).1–18 A biomechanic anatomic study by Chan and McAdams25 showed that the dorsal technique resulted in a centered screw throughout the scaphoid, whereas screws inserted volarly were centered in the waist and proximal pole but not in the distal pole. Furthermore, a hybrid technique has been described by Pirela-Cruz et al13 in which the wire is placed dorsally but the screw is placed volarly, theoretically combining the advantages and minimizing the problems of the volar and dorsal techniques. At this time, however, insufficient evidence exists to conclude which technique most effectively minimizes the risks of complications, and a randomized, controlled trial is likely needed to make a direct comparison. A final factor in the discrepancy of reported complication rates is that the definition of complication also varies considerably in the literature. For example, some researchers8 chose not to label hardware removal as a complication—an event that our methods would have called major. In another study,18 the researchers had a case of nonunion treated with a bone stimulator, but they did not specifically label this case as a complication. Our report objectively identified possible complications a priori, then grouped them into classifications based on a modification of a wrist-specific description in the existing literature.22 In previous reports in which this was not the case, complications may have gone underreported, or at least underemphasized. A lack of standardization in complication reporting makes the comparison of complication rates across studies and procedures difficult at best, and it can lead to notable differences in reported complication rates. Our study has a few limitations worth noting. The retrospective design, by nature, prohibited a more standardized study protocol. Our small sample size limited the power of the study for more in-depth
statistical analysis; however, our sample size was larger than those of many existing studies.4,9,10,26,27 Another potential limitation was the use of plain radiographs as the determining factor for assessing fracture healing. The retrospective nature of this study and patient compliance factors prevented us from being able to obtain postoperative CT scans on every patient, leaving plain radiographs as our only option for judging healing. A recent study of the radiographic diagnosis of scaphoid fractures by Cheung et al,23 however, found that all fractures were clearly visible on plain radiographs when at least 4 views were used together. Their study provides some support for our methods, as does the fact that every CT scan obtained in our series agreed with the plain radiographic determinations. Our study showed a higher complication rate than those previously reported for percutaneous cannulated screw fixation of nondisplaced scaphoid waist fractures with a dorsal technique. Complications can represent a considerable problem for surgeons and patients, and surgeons should not assume—as the current literature may encourage them to do—that all patients will have excellent results with minimal risk for problems when a percutaneous technique is used. Our report of clearly defined complications can better inform the dialogue between surgeons and patients regarding dorsal percutaneous screw fixation of scaphoid waist fractures. Received for publication December 5, 2006; accepted in revised form April 11, 2007. 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. The authors would like to thank Laura McCoy, Jean Godwin, and Cathy Wells for their assistance with this research project. Corresponding author: Brandon D. Bushnell, MD, University of North Carolina Hospitals, Department of Orthopaedic Surgery, 101 Manning Dr, CB #7055, Chapel Hill, NC 27599-7055; e-mail: [email protected]. Copyright © 2007 by the American Society for Surgery of the Hand 0363-5023/07/32A06-0010$32.00/0 doi:10.1016/j.jhsa.2007.04.003
References 1. Woazasek GE, Moser KD. Percutaneous screw fixation for fractures of the scaphoid. J Bone Joint Surg 1991;73B:138 – 142. 2. Haddad FS, Goddard NJ. Acute percutaneous scaphoid fixation: a pilot study. J Bone Joint Surg 1998;80B:95–99. 3. Adolfsson L, Lindau T, Arner M. Acutrak screw fixation versus cast immobilisation for undisplaced scaphoid waist fractures. J Hand Surg 2001;26B:192–195. 4. Bond CD, Shin AY, McBride MT, Dao KD. Percutaneous screw fixation or cast immobilization for nondisplaced scaphoid fractures. J Bone Joint Surg 2001;83A:483– 488.
Bushnell, McWilliams, and Messer / Scaphoid Complications 5. Saeden B, Tornkvist H, Ponzer S, Hoglund M. Fracture of the carpal scaphoid. A prospective, randomised 12-year follow-up comparing operative and conservative treatment. J Bone Joint Surg 2001;83B:230 –234. 6. Slade JF III, Grauer JN, Mahoney JD. Arthroscopic reduction and percutaneous fixation of scaphoid fractures with a novel dorsal technique. Orthop Clin North Am 2001;32:247–261. 7. Slade JF III, Gutow AP, Geissler WB. Percutaneous internal fixation of scaphoid fractures via an arthroscopically assisted dorsal approach. J Bone Joint Surg 2002;84A:21–36. 8. Yip HSF, Wu SC, Chang RY, So TY. Percutaneous cannulated screw fixation of acute scaphoid waist fracture. J Hand Surg 2002;27B:42– 46. 9. Jeon IH, Oh CW, Park BC, Ihn JC, Kim PT. Minimal invasive percutaneous Herbert screw fixation in acute unstable scaphoid fracture. Hand Surg 2003;8:213–218. 10. Chen AC, Chao EK, Hung SS, Lee MS, Ueng SW. Percutaneous screw fixation for unstable scaphoid fractures. J Trauma 2005;59:184 –187. 11. Dias JJ, Wildin CJ, Bhowal B, Thompson JR. Should acute scaphoid fractures be fixed? A randomized controlled trial. J Bone Joint Surg 2005;87A:2160 –2168. 12. Drac P, Manak P, Labonek I. Percutaneous osteosynthesis versus cast immobilisation for the treatment of minimally and non-displaced scaphoid fractures. Functional outcomes after a follow-up of at least 12 month. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 2005;149: 149 –151. 13. Pirela-Cruz MA, Battista V, Burnette S, Hansen T. A technical note on percutaneous scaphoid fixation using a hybrid technique. J Orthop Trauma 2005;19:570 –573. 14. Shih JT, Lee HM, Hou YT, Tan CM. Results of arthroscopic reduction and percutaneous fixation for acute displaced scaphoid fractures. Arthroscopy 2005;21:620 – 626. 15. Slade JF III, Taksali S, Safanda J. Combined fractures of the scaphoid and distal radius: a revised treatment rationale using percutaneous and arthroscopic techniques. Hand Clin 2005;21:427– 441. 16. Wong TC, Yip TH, Wu WC. Carpal ligament injuries with
17.
18.
19.
20. 21.
22.
23. 24.
25.
26.
27.
833
acute scaphoid fractures—a combined wrist injury. J Hand Surg 2005;30B:415– 418. Arora R, Gschwentner M, Krappinger D, Lutz M, Blauth M, Gabl M. Fixation of nondisplaced scaphoid fractures: making treatment cost effective: prospective controlled trial. Arch Orthop Trauma Surg 2007;127:39 – 46. Bedi A, Jebson PJL, Hayden RJ, Jacobson JA, Martus JE. Internal fixation of acute, nondisplaced scaphoid waist fractures via a limited dorsal approach: an assessment of radiographic and functional outcomes. J Hand Surg 2007;32A: 326 –333. DiMarcantonio T. Taking it back: surgeons retract podium statements to reflect new practices. Orthopedics Today 2007; 27:90. Slade JF III, Jaskwhich D. Percutaneous fixation of scaphoid fractures. Hand Clin 2001;17:553–574. Slade JF III, Geissler WB, Gutow AP, Merrell GA. Percutaneous internal fixation of selected scaphoid nonunions with an arthroscopically assisted dorsal approach. J Bone Joint Surg 2003;85A:20 –32. Beredjiklian PK, Bozentka DJ, Leung YL, Monaghan BA. Complications of wrist arthroscopy. J Hand Surg 2004;29A: 406 – 411. Cheung GC, Lever CJ, Morris AD. X-ray diagnosis of acute scaphoid fractures. J Hand Surg 2006;31B:104 –109. Streli R. [Percutaneous screwing of the navicular bone of the hand with a compression drill screw (a new method)]. Zentralbl Chir 1970;95:1060 –1078. Chan KW, McAdams TR. Central screw placement in percutaneous screw scaphoid fixation: a cadaveric comparison of proximal and distal techniques. J Hand Surg 2004:29A: 74 –79. Ledoux P, Chahidi N, Moermans J, Kinnen L. [Percutaneous Herbert screw osteosynthesis of the scaphoid bone]. Acta Orthop Belg 1995;61:43– 47. Haddad FS, Goddard NJ. Acute percutaneous scaphoid fixation using a cannulated screw. Chir Main 1998;17:119 – 126.
Purpose: With advances in tools and techniques, percutaneous screw fixation of nondisplaced fractures of the scaphoid waist has gained increasing popularity in recent years as an alternative to prolonged cast immobilization or open reduction and internal fixation. Many reports cite low complication rates, including no complications in some series. The purpose of this study was to evaluate the complications encountered with dorsal percutaneous cannulated screw fixation of nondisplaced scaphoid waist fractures. Methods: A retrospective chart review was performed for 24 patients who had surgery performed by a single surgeon over a 5-year period. All cases involved dorsal percutaneous cannulated screw fixation of nondisplaced (⬍1 mm) fractures of the scaphoid waist. Complications were rated a priori as major or minor based on modifications of established criteria. Results: The overall complication rate was 29%; there were 21% (5/24) major complications and 8% (2/24) minor complications. Major complications consisted of 1 case of nonunion, 3 cases involving hardware problems, and 1 case of postoperative fracture of the proximal pole of the scaphoid. Minor complications included intraoperative equipment breakage— 1 case involving a screw and 1 case involving a guide wire. Conclusions: Complications in dorsal percutaneous cannulated screw fixation of scaphoid fractures may be more common than previously reported. (J Hand Surg 2007;32A:827– 833. Copyright © 2007 by the American Society for Surgery of the Hand.) Type of study/level of evidence: Therapeutic IV. Key words: Scaphoid, complications, scaphoid fractures, percutaneous screw fixation, percutaneous technique, dorsal percutaneous scaphoid technique.
istorically, nondisplaced fractures of the scaphoid waist were often managed with immobilization as the standard of treatment. Recently, there has been an increased trend toward surgical fixation of nondisplaced fractures of the scaphoid with percutaneous techniques. Several reports of successful results—most of them using a volar approach— have appeared in the literature. Many of these series have reported low complication rates (Table 1).1–18 In particular, there have been 4 reports6,7,15,18 involving the dorsal approach, with reported ranges of complications from 0% to 14%. Advances in tools and techniques, coupled with a desire to eliminate the long periods of immobilization associated with nonsurgical treatment, have con-
H
tributed to the popularity of the percutaneous technique. Some surgeons have even suggested that percutaneous treatment should be considered the new standard of care for this injury.19 The purpose of this study was to evaluate and report the complications encountered with dorsal percutaneous cannulated screw fixation of nondisplaced scaphoid waist fractures.
Materials and Methods After obtaining appropriate institutional research review board approval, a retrospective review of medical records was performed for all patients who had surgical fixation of any scaphoid fracture by the senior hand surgeon at our institution (T.M.M.), an The Journal of Hand Surgery
827
828
The Journal of Hand Surgery / Vol. 32A No. 6 July–August 2007
Table 1. Selected Reports on Percutaneous Fixation of Scaphoid Fractures Study
Year
N
Union Rate, % (n)
Woazasek and Moser1 Haddad and Goddard2 Adolfsson et al3
1991 1998
146 15
89 (130) 100 (15)
Volar Volar
2001 2001 2001 2001 2002 2002 2003 2005 2005 2005 2005 2005 2005
23 11 32 16 27 49 13 11 44 26 51 15 7
96 100 100 100 100 100 100 100 100 96 98 100 100
Volar Volar Volar Dorsal Dorsal Volar Volar Volar Volar Volar Hybrid Volar Dorsal
2005
52
96 (50)
Volar
8 (4)
2006 2007
21 18
95 (20) 94 (17)
Volar Dorsal
19 (4) 6 (1)
—
577
96%
—
8%
Bond et al4 Saeden et al5 Slade et al6 Slade et al7 Yip et al8 Jeon et al9 Chen et al10 Dias et al11 Drac et al12 Pirela-Cruz et al13 Shih et al14 Slade et al15 Wong et al16
Approach
(22) (11) (32) (16) (27) (49) (13) (11) (44) (25) (50) (15) (7)
Complication Rate, % (n) 9 (13) 7 (1) 8 9 6 0 0 0 8 9 30 4 2 0 14
(2) (1) (2) (0) (0) (0)* (1) (1) (13) (1) (1) (0) (1)
Arora et al17 Bedi et al18 Totals/average
Complications See below Radial Nerve Dysesthesia Nonunion, reflex sympathetic dystrophy Prominent Hardware See below None None None* RSD Superficial skin infection See below Nonunion Nonunion None EPL Rupture⫹ Nonunion (2), Delayed Union (2) Nonunion, superficial wound infection, CRPS (2) Nonunion Nonunion most common complication
The numbers included represent (when specified by the researchers) acute fractures with full follow-up evaluation: Woazasek and Moser1: nonunion (5), superficial radial nerve irritation (3), superficial wound infection (2), Sudeck’s atrophy (2), loose/painful screw (1). Saeden et al5: loose screw (1), failure of the screw to enter the proximal fragment (1). Dias et al11: hypertrophic scar (4), sensitive and hypertrophic scar (3), sensitive scar (3), superficial wound infection (1), hypoesthesia in region of palmar cutaneous branch of median nerve (1), mild early algodystrophy (1). † The extensor pollicis longus rupture was due to a dorsal plate used to treat a concomitant ipsilateral distal radius fracture. *These researchers did not consider hardware removal a postoperative complication.
academic-affiliated level I trauma center. Inclusion criteria included a nondisplaced (⬍1 mm) fracture of the scaphoid waist, a time from injury of less than 10 weeks, treatment involving dorsal percutaneous screw fixation as described by Slade et al,6,7,15,20,21 and at least a follow-up period of at least 3 months. Exclusion criteria were initial fracture of the proximal or distal pole, fracture displacement of more than 1 mm, carpal instability, a delayed diagnosis of longer than 10 weeks, the presence of an established nonunion or malunion, use of open technique, and a follow-up period of less than 3 months. Concomitant injuries were not considered exclusion criteria. Within the 5-year period between October 1, 2001 and September 30, 2006, 24 scaphoid fractures in 24 patients met our criteria and were included in this study. The average follow-up interval for these patients was 9 months (range, 3–24 mo). The average age of these 24 patients at the time of surgery was 28 years (range, 14 –53 y). Fifteen surgeries were performed on the left hand and 9 on the right. The
dominant hand was involved in 12 (50%) of the patients. Twenty-three cases involved the Accutrak screw system (Acumed, Inc., Beaverton, OR), and 1 case involved the TwinFix screw system (Stryker Leibinger Inc., Kalamazoo, MI). The primary data points for this study were the incidence and type of complications. We classified both intraoperative and postoperative complications into major and minor groups, based on a modified version of the method of the assessment of complications in wrist arthroscopy described by Beredjiklian et al.22 Major complications included any complication resulting in the need for additional surgical intervention, nonunion, malunion, fracture, compartment syndrome, permanent nerve injury, septic arthritis, vascular injury, complex regional pain syndrome, permanent stiffness, or tendon rupture. Minor complications included any complication that required additional nonsurgical intervention, resulted in deviation from the standard surgical technique, prolonged the surgical time, led to superficial
Bushnell, McWilliams, and Messer / Scaphoid Complications
829
infection or prolonged wound drainage (⬎5 days), or caused transient neurologic symptoms, stiffness, or tendon irritation. Because our definition of complications included nonunions and malunions, radiographic healing was assessed and defined as the lack of a visible radiolucent fracture line on any one of a minimum of 4 views of the scaphoid, as judged by the authors and by the radiologist’s report.23 Computed tomography (CT) scans could not be obtained for every patient and therefore were not used as the primary criterion for judging healing, but CT images were used to confirm plain radiographic findings in each patient for whom they were available.
Results Major Complications There were 5 major complications (21%), all occurring in the postoperative period: 1 case of nonunion, 3 cases involving hardware problems, and 1 case of postoperative fracture of the proximal pole of the scaphoid. The case of nonunion was diagnosed in a 38-yearold man, a smoker, 8 months after surgery for the initial injury. It was treated with screw removal and
Figure 1. Errant screw placement. A sagittal CT image shows dorsal malpositioning of the screw with inadequate capture of the distal fragment.
Figure 2. Delayed union and settling of the screw. An anteroposterior radiograph of a 30-year-old male smoker whose fracture had not healed at 11 months after percutaneous fixation. He was taken to surgery for a suspected nonunion, but the fracture was found to be rigidly healed at surgery. As shown in this radiograph, however, the screw had settled during the delayed time to union and remained prominent at the proximal pole. Screw removal resulted in resolution of the patient’s symptoms.
revision open reduction, internal fixation, and bone grafting through a volar approach. The patient then went on to heal over the course of several additional months. Three cases involved problems with the screws, necessitating an additional operation. One patient’s postoperative CT scan showed errant screw placement with dorsal malpositioning and inadequate capture of the distal fragment (Fig. 1). This patient healed after treatment with hardware removal and revision percutaneous screw fixation via a volar approach. Another patient’s screw was too long distally and caused symptomatic irritation. A third patient had a delayed union that allowed the screw to settle and serve as a source of pain (Fig. 2). These 2 patients’ symptoms resolved with removal of the screws. A unique complication occurred in a patient in whom the fracture had radiographically and clini-
830
The Journal of Hand Surgery / Vol. 32A No. 6 July–August 2007
Figure 3. Development of a proximal pole fracture. (A) A 30-year-old man had percutaneous fixation with a 20-mm Accutrak screw, and the fracture healed at 8 weeks. (B) At 11 months after surgery, however, he still complained of pain and stiffness in the wrist. Radiographs showed a fracture of the proximal pole of the scaphoid. The exact cause of this fracture remains unclear.
cally healed at 8 weeks after percutaneous screw fixation (Fig. 3A); the patient enjoyed several months of asymptomatic, normal function. At approximately 10 months after surgery, he began to have gradually worsening pain and stiffness despite no additional trauma to the wrist. Radiographs 11 months after surgery showed a fracture of the proximal pole of the scaphoid (Fig. 3B). Arthroscopic surgery to evaluate the proximal scaphoid and hardware removal was offered, but the patient instead requested nonsurgical immobilization to attempt to address his symptoms. Minor Complications There were 2 minor complications (8%), both occurring during surgery and both rated as minor because they involved deviation from standard technique and prolonged the surgery time.22 In the 1 case involving the use of the TwinFix screw system (Stryker Leibinger Inc.), the screw broke during insertion via separation of the proximal screw head from the screw shaft. An extensive effort was needed to remove the partially inserted broken shaft, followed by conversion to a larger
screw size. Another case involved breakage of a guidewire during retrograde passage, resulting in approximately 15 minutes of prolonged surgical time for wire removal (Fig. 4).
Figure 4. Intraoperative breakage of the guidewire. Failure to maintain the wrist in flexion can lead to breakage of the guidewire. In this case, the wire impinged on the dorsal wrist capsule during readvancement of the wire back out of the dorsal entry wound. Slight flexion of the wrist then broke the wire. The small wire fragment required approximately 15 minutes of additional surgical time to remove.
Bushnell, McWilliams, and Messer / Scaphoid Complications
Discussion Percutaneous treatment of scaphoid fractures has long been investigated as an alternative to open techniques. Streli24 first described the use of the percutaneous technique in 1970. Since that time, many additional reports have described the use of percutaneous fixation of scaphoid fractures, including several recent series1–18 with low complication rates (Table 1). Our results show higher complication rates than most of these contemporary reports,1–18 especially those using the dorsal technique. Complications reported for the dorsal technique range from 0% to 14%, yet we report a total rate of 29%.6,7,15,18 In 4 previous studies, 6,7,15,18 only 2 complications were reported out of 67 total patients, representing an overall rate of 3%. One reported complication consisted of an extensor pollicis longus tendon rupture secondary to irritation from a dorsal plate used to treat a concomitant ipsilateral distal radius fracture— essentially a problem not related at all to the treatment of the scaphoid fracture.15 The other reported complication consisted of a case of nonunion18—a problem also seen in our study and the most common complication overall in all reports of the percutaneous technique using any approach (Table 1).1–18 There are several potential reasons for the observed discrepancy in complication rates. One of the reasons is operator error. In the minor complication involving the broken guidewire, the wrist was not completely flexed to 90° during the retrograde passage of the wire, which backed out through the dorsal skin incision, resulting in wire impingement and breakage as it was advanced under power. The risk of complications such as this one can be minimized by meticulous adherence to technique. Another reason is equipment failure. In the case of the broken TwinFix screw (Stryker Leibinger Inc.), the components separated during the insertion of a screw that otherwise appeared normal. Such problems are likely unforeseeable and thus unpreventable. Some of the complications we observed may relate to the theory underlying the dorsal approach. With a dorsal approach, the radioscaphoid joint is violated by the drill, and any screw left prominent proximally can damage the joint surface of the scaphoid fossa of the radius, a key articular surface of the wrist. The newly described potential for delayed presentation of a proximal pole fracture of the scaphoid (Fig. 3) raises the issue of whether placing a drill into this known high-risk area of the bone is a good idea. Indeed, perhaps the fracture was due to unrecognized
831
avascular necrosis of the proximal pole caused by the drill. Furthermore, it is unknown what other longterm sequelae may result from making a large hole in the proximal scaphoid articular surface. Our case involving excessive distal screw length highlights another issue in the analysis of complications common to both the volar and distal techniques—the potential interaction of technique and equipment in measuring the appropriate, or perhaps inappropriate, length of the screw. We have found that the depth gauge used with the screw set routinely overestimates the required length for proper screw placement. A recent study by Bedi et al18 advocated subtracting 4 mm from the measured length. Similarly, we have found it necessary to subtract 6 mm from the given depth gauge length and in addition to confirm this number using a parallelguidewire technique (Fig. 5). The case of the proximal pole fracture highlights the danger of erring too far in the other direction and underestimating screw length (Fig. 3). In this case, it could be that the stress-riser created by the short screw led to the development of a fracture and worsened symptoms. Achieving an exact appropriate screw length within the bone—not too short, but not too long—thus appears to be critical to success and decreased postoperative complications. In cases involving tapered screw design, such as the Accutrak system (Acumed, Inc.) used in most of our patients, it becomes even more critical to measure screw length appropriately. If a screw’s excess
Figure 5. Use of an extra guidewire to confirm length. A second guidewire of equal length is placed directly parallel to the first wire to confirm the measured length for the screw. The portion of the second wire that extends beyond the first wire equals the interosseous portion of the first wire and thus should correlate with the measurement obtained from the depth gauge.
832
The Journal of Hand Surgery / Vol. 32A No. 6 July–August 2007
length necessitates replacement with a shorter screw, the initial interfragmentary compression and overall construct rigidity are considerably reduced because of the size mismatch between the second screw’s threads and the original path in the bone. Frequent fluoroscopic images during screw insertion can help the surgeon predict excess length, so that a smaller screw can be substituted if needed before completely seating the first screw. These potential issues with the dorsal technique raise the question of which percutaneous approach (dorsal/ proximal/antegrade or volar/distal/retrograde) is the best choice for minimizing complications. Complication rates ranging from 0% to 30% have been reported in series using a volar approach (Table 1).1–18 A biomechanic anatomic study by Chan and McAdams25 showed that the dorsal technique resulted in a centered screw throughout the scaphoid, whereas screws inserted volarly were centered in the waist and proximal pole but not in the distal pole. Furthermore, a hybrid technique has been described by Pirela-Cruz et al13 in which the wire is placed dorsally but the screw is placed volarly, theoretically combining the advantages and minimizing the problems of the volar and dorsal techniques. At this time, however, insufficient evidence exists to conclude which technique most effectively minimizes the risks of complications, and a randomized, controlled trial is likely needed to make a direct comparison. A final factor in the discrepancy of reported complication rates is that the definition of complication also varies considerably in the literature. For example, some researchers8 chose not to label hardware removal as a complication—an event that our methods would have called major. In another study,18 the researchers had a case of nonunion treated with a bone stimulator, but they did not specifically label this case as a complication. Our report objectively identified possible complications a priori, then grouped them into classifications based on a modification of a wrist-specific description in the existing literature.22 In previous reports in which this was not the case, complications may have gone underreported, or at least underemphasized. A lack of standardization in complication reporting makes the comparison of complication rates across studies and procedures difficult at best, and it can lead to notable differences in reported complication rates. Our study has a few limitations worth noting. The retrospective design, by nature, prohibited a more standardized study protocol. Our small sample size limited the power of the study for more in-depth
statistical analysis; however, our sample size was larger than those of many existing studies.4,9,10,26,27 Another potential limitation was the use of plain radiographs as the determining factor for assessing fracture healing. The retrospective nature of this study and patient compliance factors prevented us from being able to obtain postoperative CT scans on every patient, leaving plain radiographs as our only option for judging healing. A recent study of the radiographic diagnosis of scaphoid fractures by Cheung et al,23 however, found that all fractures were clearly visible on plain radiographs when at least 4 views were used together. Their study provides some support for our methods, as does the fact that every CT scan obtained in our series agreed with the plain radiographic determinations. Our study showed a higher complication rate than those previously reported for percutaneous cannulated screw fixation of nondisplaced scaphoid waist fractures with a dorsal technique. Complications can represent a considerable problem for surgeons and patients, and surgeons should not assume—as the current literature may encourage them to do—that all patients will have excellent results with minimal risk for problems when a percutaneous technique is used. Our report of clearly defined complications can better inform the dialogue between surgeons and patients regarding dorsal percutaneous screw fixation of scaphoid waist fractures. Received for publication December 5, 2006; accepted in revised form April 11, 2007. 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. The authors would like to thank Laura McCoy, Jean Godwin, and Cathy Wells for their assistance with this research project. Corresponding author: Brandon D. Bushnell, MD, University of North Carolina Hospitals, Department of Orthopaedic Surgery, 101 Manning Dr, CB #7055, Chapel Hill, NC 27599-7055; e-mail: [email protected]. Copyright © 2007 by the American Society for Surgery of the Hand 0363-5023/07/32A06-0010$32.00/0 doi:10.1016/j.jhsa.2007.04.003
References 1. Woazasek GE, Moser KD. Percutaneous screw fixation for fractures of the scaphoid. J Bone Joint Surg 1991;73B:138 – 142. 2. Haddad FS, Goddard NJ. Acute percutaneous scaphoid fixation: a pilot study. J Bone Joint Surg 1998;80B:95–99. 3. Adolfsson L, Lindau T, Arner M. Acutrak screw fixation versus cast immobilisation for undisplaced scaphoid waist fractures. J Hand Surg 2001;26B:192–195. 4. Bond CD, Shin AY, McBride MT, Dao KD. Percutaneous screw fixation or cast immobilization for nondisplaced scaphoid fractures. J Bone Joint Surg 2001;83A:483– 488.
Bushnell, McWilliams, and Messer / Scaphoid Complications 5. Saeden B, Tornkvist H, Ponzer S, Hoglund M. Fracture of the carpal scaphoid. A prospective, randomised 12-year follow-up comparing operative and conservative treatment. J Bone Joint Surg 2001;83B:230 –234. 6. Slade JF III, Grauer JN, Mahoney JD. Arthroscopic reduction and percutaneous fixation of scaphoid fractures with a novel dorsal technique. Orthop Clin North Am 2001;32:247–261. 7. Slade JF III, Gutow AP, Geissler WB. Percutaneous internal fixation of scaphoid fractures via an arthroscopically assisted dorsal approach. J Bone Joint Surg 2002;84A:21–36. 8. Yip HSF, Wu SC, Chang RY, So TY. Percutaneous cannulated screw fixation of acute scaphoid waist fracture. J Hand Surg 2002;27B:42– 46. 9. Jeon IH, Oh CW, Park BC, Ihn JC, Kim PT. Minimal invasive percutaneous Herbert screw fixation in acute unstable scaphoid fracture. Hand Surg 2003;8:213–218. 10. Chen AC, Chao EK, Hung SS, Lee MS, Ueng SW. Percutaneous screw fixation for unstable scaphoid fractures. J Trauma 2005;59:184 –187. 11. Dias JJ, Wildin CJ, Bhowal B, Thompson JR. Should acute scaphoid fractures be fixed? A randomized controlled trial. J Bone Joint Surg 2005;87A:2160 –2168. 12. Drac P, Manak P, Labonek I. Percutaneous osteosynthesis versus cast immobilisation for the treatment of minimally and non-displaced scaphoid fractures. Functional outcomes after a follow-up of at least 12 month. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 2005;149: 149 –151. 13. Pirela-Cruz MA, Battista V, Burnette S, Hansen T. A technical note on percutaneous scaphoid fixation using a hybrid technique. J Orthop Trauma 2005;19:570 –573. 14. Shih JT, Lee HM, Hou YT, Tan CM. Results of arthroscopic reduction and percutaneous fixation for acute displaced scaphoid fractures. Arthroscopy 2005;21:620 – 626. 15. Slade JF III, Taksali S, Safanda J. Combined fractures of the scaphoid and distal radius: a revised treatment rationale using percutaneous and arthroscopic techniques. Hand Clin 2005;21:427– 441. 16. Wong TC, Yip TH, Wu WC. Carpal ligament injuries with
17.
18.
19.
20. 21.
22.
23. 24.
25.
26.
27.
833
acute scaphoid fractures—a combined wrist injury. J Hand Surg 2005;30B:415– 418. Arora R, Gschwentner M, Krappinger D, Lutz M, Blauth M, Gabl M. Fixation of nondisplaced scaphoid fractures: making treatment cost effective: prospective controlled trial. Arch Orthop Trauma Surg 2007;127:39 – 46. Bedi A, Jebson PJL, Hayden RJ, Jacobson JA, Martus JE. Internal fixation of acute, nondisplaced scaphoid waist fractures via a limited dorsal approach: an assessment of radiographic and functional outcomes. J Hand Surg 2007;32A: 326 –333. DiMarcantonio T. Taking it back: surgeons retract podium statements to reflect new practices. Orthopedics Today 2007; 27:90. Slade JF III, Jaskwhich D. Percutaneous fixation of scaphoid fractures. Hand Clin 2001;17:553–574. Slade JF III, Geissler WB, Gutow AP, Merrell GA. Percutaneous internal fixation of selected scaphoid nonunions with an arthroscopically assisted dorsal approach. J Bone Joint Surg 2003;85A:20 –32. Beredjiklian PK, Bozentka DJ, Leung YL, Monaghan BA. Complications of wrist arthroscopy. J Hand Surg 2004;29A: 406 – 411. Cheung GC, Lever CJ, Morris AD. X-ray diagnosis of acute scaphoid fractures. J Hand Surg 2006;31B:104 –109. Streli R. [Percutaneous screwing of the navicular bone of the hand with a compression drill screw (a new method)]. Zentralbl Chir 1970;95:1060 –1078. Chan KW, McAdams TR. Central screw placement in percutaneous screw scaphoid fixation: a cadaveric comparison of proximal and distal techniques. J Hand Surg 2004:29A: 74 –79. Ledoux P, Chahidi N, Moermans J, Kinnen L. [Percutaneous Herbert screw osteosynthesis of the scaphoid bone]. Acta Orthop Belg 1995;61:43– 47. Haddad FS, Goddard NJ. Acute percutaneous scaphoid fixation using a cannulated screw. Chir Main 1998;17:119 – 126.