- Email: [email protected]
Variation in Digital Rotation and Alignment in Normal Subjects
SCIENTIFIC ARTICLE
Variation in Digital Rotation and Alignment in Normal Subjects Virak Tan, MD, Tosca Kinchelow, MD, Pedro K. Beredjiklian, MD
Purpose The purpose of this study is to quantify finger/digit rotation, overlap, parallelism, and convergence to the scaphoid tuberosity in normal volunteers to establish standards for comparison. Methods We examined 240 uninjured fingers in 30 volunteers. There were 14 men and 16 women with an average age of 35 years. Rotation was determined with the palm flat and fingers extended. Digit overlap, parallelism (angular relationship between the index, ring, and small fingers with respect to the middle finger) and scaphoid convergence were determined with simultaneous flexion of metacarpophalangeal and proximal interphalangeal joints. Linear and angular measurements were performed with imaging software. Results Rotation, parallelism, and scaphoid convergence measurements were similar comparing left with right hands. Rotation: All digits were found to be in supination relative to the horizontal plane; the small fingers averaged 9°, the ring fingers 4°, the middle fingers 8°, and the index fingers 12°. Parallelism: Angular measurements between the middle finger and the small finger averaged 19°, between the middle finger and the ring finger 9°, and between the middle finger and the index finger 11°. Scaphoid convergence: In no hand did all 4 fingers converge onto scaphoid tuberosity. Whereas small and ring finger trajectories averaged ⫺0.1 to 1.2 mm from the scaphoid tubercle, the middle finger averaged 4.0 to 4.6 mm and the index finger 8.1 to 9.5 mm. Overlap: Ninety percent of individuals demonstrated digit overlap, although none covered more than 50% of the adjacent nail plate. Seventy-seven percent of these were bilateral, and 73% were asymmetric. Conclusions This study establishes the normal parameters for digit rotation, overlap, parallelism, and scaphoid convergence. For digit rotation, parallelism, and scaphoid convergence, the contralateral (uninjured) hand can be used reliably for comparison. However, for digit overlap, the contralateral hand should not be used for comparison because of side-to-side asymmetry and variability. (J Hand Surg 2008;33A:873–878. Copyright © 2008 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Diagnostic IV. Key words Digit alignment, digit overlap, digit rotation, finger fracture.
From the Department of Orthopaedics, Division of Hand, Upper Extremity and Microvascular Surgery, UMDNJ-NJ Medical School, Newark, NJ; and the Department of Orthopaedic Surgery, Penn Hand Center, University of Pennsylvania Health System, Philadelphia, PA. Received for publication November 15, 2007; accepted in revised form February 6, 2008. No benefits in any form have been received or will be received related directly or indirectly to the subject of this article. Corresponding author: Virak Tan, MD, Department of Orthopaedics, Division of Hand, Upper Extremity and Microvascular Surgery, UMDNJ-NJ Medical School, 140 Bergen St. (ACC South, D Level, Orthopaedics), Newark, NJ 07103; e-mail: [email protected]. 0363-5023/08/33A06-0011$34.00/0 doi:10.1016/j.jhsa.2008.02.006
ETACARPAL AND PHALANGEAL fractures are common, representing more than 40% of all hand and forearm fractures.1 Although most of these injuries can be successfully treated nonoperatively, certain fractures are better managed with surgery. It is essential to properly assess fracture pattern and stability, as this will guide fracture management. Undertreatment of an unstable fracture can result in deformity and functional limitation.2–5 One of the more ambiguous parameters to measure in a fractured finger is rotational deformity. Various techniques for mea-
M
© ASSH 䉬 Published by Elsevier, Inc. All rights reserved. 䉬 873
874
DIGITAL ROTATION AND ALIGNMENT IN NORMAL SUBJECTS
FIGURE 2: Special hand positioning device that passively holds the MCP and PIP joints in flexion. FIGURE 1: Photograph of the hand in position 1, with the palm flat on a table with fingers extended. The photograph is centered on the middle finger. The rotation of the digit is established by measuring the angle between the nail plate and the horizontal surface (shown for the middle finger).
surement and acceptable parameters have been described,2,4,6,7 and most clinicians determine malrotation of finger fractures by comparison with the uninjured hand. Assuming that there is minimal side-to-side variability, most clinicians determine malrotation of finger fractures by comparison with the uninjured hand. The purpose of this study was to determine whether the aforementioned assumption is correct by quantifying side-to-side variations in digit rotation, digit overlap, and digit convergence to the scaphoid tuberosity in normal volunteers. MATERIALS AND METHODS We examined 240 fingers (index, middle, ring, and small) in 30 volunteers without prior hand injury. There were 14 men and 16 women with an average age of 35 years (range, 20 – 63 years). Digital photographs of the hands were obtained in 2 positions. Position 1 was with the palm flat on a horizontal surface, full finger extension, wrist in neutral alignment, and the camera centered down the axis of the middle finger (Fig. 1). Position 2 was obtained with the hand placed in a special passive positioning device (Fig. 2) allowing simultaneous flexion of the metacarpophalangeal (MCP) and the proximal interphalangeal (PIP) joints. The wrist was in neutral alignment and the camera centered over the interval between the middle and ring fingers (Fig. 3). Rotation was determined by the angle of the nail plate relative to a horizontal plane with the hand in position.1.2,4,6,7 Digit overlap, parallelism, angulation, and convergence to the scaphoid tuberosity were determined with the hand in position 2. Digit
FIGURE 3: Photograph of a hand in position 2, side view. Digit overlap, angulation, and convergence to the scaphoid tuberosity were determined with the hand in position 2 but the photographs were taken vertically (see Figs. 4, 5).
overlap was defined as any obscuring of the nail plate of one finger by an adjacent finger and was graded in 25% increments of the nail plate covered (Fig. 4). Parallelism was determined with respect to the middle finger and defined as the angle formed between a longitudinal trajectory of the middle finger and each of the other fingers (Fig. 5). Scaphoid tubercle convergence was assessed by marking a 1-cm circle around the scaphoid tubercle volarly and then measuring the horizontal distance between it and a line drawn along the trajectory of each digit (Fig. 5). The determination of scaphoid tubercle position was made by 1 observer, and the variability in determining this landmark was not performed. Angular and linear measurements were made on the digital photographs, and analysis was performed by a
JHS 䉬 Vol A, July–August
DIGITAL ROTATION AND ALIGNMENT IN NORMAL SUBJECTS
875
RESULTS The results for digit rotation, parallelism, and scaphoid convergence are presented in Table 1. Rotation Left side to right side digit rotation for each finger was found to be similar and was in relative supination. Rotation of the small finger averaged 8° on the left versus 10° on the right (p ⫽ .12); the ring finger, 4° versus 4° (p ⫽ .79); the middle finger, 8° versus 9° (p ⫽ .48); and the index finger, 12° versus 12° (p ⫽ .98), left versus right, respectively. FIGURE 4: Photograph showing 25% overlap of the ring finger by the middle finger. The black circle marks the palpable scaphoid tubercle volarly.
Parallelism Parallelism relative to middle finger was similar between the 2 sides. Angular measurements between the middle finger and the small finger averaged 19° on the left and 19° on the right (p ⫽ .93); between the middle finger and the ring finger, 9° versus 9° (p ⫽ .67); and between the middle finger and the index finger,10° versus 11° (p ⫽ .54), left versus right, respectively. Scaphoid tubercle convergence In no hand did all 4 fingers converge onto scaphoid tuberosity. However, there remained symmetry with side-to-side comparison; p value for left versus right in the small finger is .26, for the ring finger .06, for the middle finger .49, and for the ring finger .26. See Table 1 for ranges, standard deviations, and 95% confidence intervals for the data.
FIGURE 5: Scaphoid tubercle convergence was assessed by marking the palpable scaphoid tubercle volarly (black circle) and then measuring the horizontal distance between it and a line drawn along the trajectory of each digit. In this photograph, the middle and ring finger trajectories fall onto the tuberosity, but those of the index and small fingers do not. Parallelism is defined as the angle formed between a longitudinal trajectory of the middle finger and each of the other fingers.
single observer (V.T.) on a Macintosh computer using the public domain NIH Image program (developed at the U.S. National Institutes of Health and available on the Internet at http://rsb.info.nih.gov/ nih-image/). t-tests were performed for rotation, parallelism, and scaphoid tuberosity convergence. Chisquare analysis was used to test significance for the nominal variable in digit overlap. A p value less than .05 was considered significant.
Overlap Ninety percent of individuals had partial overlap of the nail plates of at least 1 finger. In 23 subjects, the overlap occurred in both hands. Comparing left with right hands, there was asymmetry (ie, overlap of different fingers) in 22 of 30 subjects (p ⬍ .001). In all digits that had overlap, the nail plate was covered 50% or less. DISCUSSION Indications for fixation of metacarpal and phalangeal fractures of the hand are based on angular and rotational deformity, bony apposition, comminution, and joint stability. Whereas various measurements may be considered acceptable depending on which bone is fractured and the location of the fracture (head, neck, shaft, base), even a small degree of malrotation (0 to 10°) has been an indication for fixation.2,4,6,8 One degree of rotation at the fracture site may translate to 5° at the fingertip,9 and as little as 5° of fracture malrotation can cause 1.5 cm of digital overlap when the fingers are flexed.4 Despite the importance of proper rotational
JHS 䉬 Vol A, July–August
876
TABLE 1. Normal Ranges of Digit Rotation, Parallelism, and Scaphoid Convergence Left Hand
Right Hand
SF
RF
MF
8
4
8
IF
IF
MF
RF
9
4
SF
Digit rotation* (°) Range SD JHS 䉬 Vol A, July–August
95% confidence interval
12
12
10
⫺3 to 22
⫺8 to 17
1 to 15
1 to 25
⫺2 to 25
⫺1 to 24
⫺11 to 16
0.2 to 30
5.5
5.6
4.5
5.5
6.3
6.9
6.9
7.3
5.8 to 9.9
1.7 to 5.8
6.1 to 9.4
9.9 to 14.0
9.6 to 14.4
6.2 to 11.3
1.6 to 6.8
7.7 to 13.1
p value
.12
.79
.48
.98
Digit parallelism†(°) Mean Range SD 95% confidence interval
19
9
N/A
10
11
9
19
10 to 31
2 to 20
N/A
3 to 20
N/A
0 to 17
7 to 39
5.6
4.2
N/A
4.2
4.0
N/A
4.0
6.8
17.3 to 21.2
7.6 to 10.6
N/A
8.9 to 11.9
9.5 to 12.4
N/A
7.2 to 10.0
16.7 to 21.6
—
.54 9.5
4.6
1.2
0.5
p value
.93
.67
0.3 to 18
N/A
Scaphoid convergence‡ (mm) Mean
0
⫺0.1
4.0
Range
⫺10.4 to 6.9
⫺6.2 to 5.2
⫺3.1 to 14.9
2.0
4.5
SD 95% confidence interval p value
2.8 ⫺2.5 to 2.0 .26
⫺0.5 to 3.0 .06
5.7 to 10.1 .49
8.1 0 to 22.0
0 to 21.3
0 to 17.0
⫺0.7 to 6.3
5.6
6.1
4.5
1.9
10.6 to 15.0
12.3 to 16.7
6.8 to 10.8
1.7 to 5.1
.26
SF, small finger; RF, ring finger; MF, middle finger; IF, index finger; N/A, not applicable. *With MCP and PIP joints in extension; rotation relative to horizontal plane; positive value ⫽ supination, and negative value ⫽ pronation. †With finger flexion in device; relative to axis of long finger. ‡With finger flexion in device; positive value ⫽ radial, and negative value ⫽ ulnar to scaphoid tuberosity.
⫺6.3 to 8.5 3.2 ⫺1.0 to 3.5
DIGITAL ROTATION AND ALIGNMENT IN NORMAL SUBJECTS
Mean
DIGITAL ROTATION AND ALIGNMENT IN NORMAL SUBJECTS
alignment, it has been relatively difficult to assess reproducibly.10 Several methods have been used. With MCP and PIP joint flexion, increased finger overlap or scissoring (compared with the contralateral side) indicates malrotation.4 Furthermore, it has been reported that in this position, the fingertips normally point to the scaphoid tuberosity.2,4 Soft tissue swelling alone can cause apparent malrotation. Smith et al.7 simulated soft tissue swelling by injecting saline into the fourth web spaces of noninjured volunteers and assessed digital rotation radiographically and clinically in flexion and extension. After injection, all subjects demonstrated a new onset of digital overlap of the flexed small finger on the ring finger (clinically and radiographically). However, none demonstrated malrotation in extension, and the angles between the nail plates and the horizontal plane remained stable. They concluded that small finger internal rotation/supination (pseudorotation) can be associated with fourth web space edema and could cause falsely increased scissoring/overlap in flexion. In our study, 27 individuals had overlap of the fingers. Although 23 of these were bilateral, 22 were asymmetric in side-to-side comparison. This suggests that comparing the injured side to the uninjured side may not be reliable when assessing digit overlap (scissoring) with finger flexion. Based on our findings, we propose that 50% or less overlap of the nail plate in any digit is within normal limits and does not cause functional difficulties (“50% rule”). It has been written that all digits should point to the scaphoid tubercle when in flexion (ie, scaphoid convergence test); however, there is a discrepancy in how this test should be performed. Some published reports illustrate the maneuver with all fingers flexed simultaneously,2,4 while others had the digits flexed individually.11,12 We performed the test with all the fingers simultaneously flexed, and none of the patients demonstrated convergence of all fingertips to the scaphoid tubercle (Fig. 5). As such the concept of “scaphoid convergence” may only hold true when the digits are flexed individually. We did not identify any statistical differences from side to side for each finger when evaluating scaphoid tubercle convergence or parallelism. Because of this side-to-side symmetry, we believe that comparison of convergence and parallelism to the contralateral hand is a more accurate determination of rotation in those cases where the MCPs and PIPs can be simultaneously flexed. In addition, parallelism measurements can be
877
done independent of the location of the scaphoid tubercle. Another method for assessing digit rotation is measuring the angle of the nail plate relative to the horizontal plane, with the palm flat and fingers extended.2,4,6,7,13 This method obviates the need to try to flex all the fingers down. Our data challenge the notion that digit rotation within 10° of neutral is within normal limits.6,14 We found that all digits were in relative supination with the index finger having an average of 12°, and the range for the small finger was up to 30°. Based on our data, normal digit rotation for the small fingers is 9° ⫾ 6, the ring fingers 4° ⫾ 6, the middle fingers 8° ⫾ 6, and the index fingers 12° ⫾ 6. In our subjects, the average rotation for each digit is within 5° of the findings of Bansal and Craigen.13 The current study establishes the normal ranges of rotation for each digit and side-to-side similarity, which can aid in proper assessment of rotation (Table 1). In a similar study, Bansal and Craigen13 describe the normal parameters of digital rotation based on the assessment of fingernail orientation with the fingers in full extension and with the MCP joints fully flexed. These authors conclude that the orientation of the fingernails can vary within and between individuals. In our study, digit alignment was also measured with the digits fully flexed, in a manner that mimics the clinical setting more accurately and addresses the main concern with malrotation, which is digital overlap.15 We demonstrate that although in most circumstances comparison with the contralateral (uninjured) side may be sufficient, digit overlap is often asymmetric. Side-to-side comparison in this case may be misleading, but clinicians can use the “50% rule” as a guide. Furthermore, in the instance of a previous or acute contralateral hand injury, knowledge of normal ranges can be very helpful. With critical evaluation of normal anatomic alignment, our data challenge the widely accepted teaching that all fingertips point to the scaphoid tuberosity on attempted fist formation. A limitation of our study is that we did not have a comparison group with fractures. However, the purpose of the study was to establish normal values to be the baseline when assessing finger alignment. A second limitation is that the reference landmarks were determined by palpation (of the scaphoid tuberosity) and visual inspection, which could have been subjected to bias or errors. However, we believed that this represented the real-life clinical situation when trying to assess finger rotation and alignment. An additional limitation is the lack of reliability coefficients for
JHS 䉬 Vol A, July–August
878
DIGITAL ROTATION AND ALIGNMENT IN NORMAL SUBJECTS
the data because we only had a single observer. Finally, in the absence of any clinical data, we are unable to determine what degree of digital malrotation is clinically important. REFERENCES 1. Henry MH. Fractures and dislocations of the hand. In: Bucholz RW, ed. Rockwood & Green’s fractures in adults. 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2006:772. 2. Ashkenaze DM, Ruby LK. Metacarpal fractures and dislocations. Orthop Clin North Am 1992;23:19 –33. 3. Diao E, Welborn JH. Extraarticular fractures of the metacarpals. In: Berger R, Weiss APC, eds. Hand surgery. 1st ed. Philadelphia: Lippincott Williams & Wilkins, 2004:139 –151. 4. Jupiter JB, Axelrod TS, Belsky MR. Fractures and dislocations of the hand. In: Browner BB, ed. Skeletal trauma. 3rd ed. Philadelphia: Saunders, 2003:1153. 5. Stern PJ. Fractures of the metacarpals and phalanges. In: Green D, Hotchkiss R, Pederson W, Wolfe S, eds. Green’s operative hand surgery. 5th ed. Philadelphia: Elsevier, 2005:277–341.
6. Royle SG. Rotational deformity following metacarpal fracture. J Hand Surg 1990;15B:124 –125. 7. Smith NC, Moncrieff NJ, Hartnell N, Ashwell J. Pseudorotation of the little finger metacarpal. J Hand Surg 2003;28B:395–398. 8. Barton NJ. Fractures of the hand. J Bone Joint Surg 1984;66B:159 – 167. 9. Opgrande JD, Westphal SA. Fractures of the hand. Orthop Clin North Am 1983;14:779 –792. 10. Bain GI, Guy SD. Derotation taping for metacarpal and phalangeal fractures. Tech Hand Up Extrem Surg 2003;7:130 –132. 11. Smith P. Lister’s the hand: diagnosis and indications, 4th ed. New York: Churchill Livingstone, 2002: 13. 12. Chang J. Plastic surgery: contributions to hand surgery. In: Hentz V, ed. Mathes’ plastic surgery, volume VII: the hand and upper limb, part 1, 2nd ed. Philadelphia: Elsevier, 2006:20. 13. Bansal R, Craigen MAC. Rotational alignment of the finger nails in a normal population. J Hand Surg [Eur] 2007;32:80 – 84. 14. Smith RJ and Peimer CA. Injuries to the metacarpal bones and joints. Adv Surg 1977;11:341–374. 15. Rosenwasser MP, Quitkin HM. Malunion and other posttraumatic complications in the hand. In: Berger R, Weiss APC, eds. Hand surgery, 1st ed. Philadelphia: Lippincott Williams & Wilkins, 2004: 207–231.
JHS 䉬 Vol A, July–August
Variation in Digital Rotation and Alignment in Normal Subjects Virak Tan, MD, Tosca Kinchelow, MD, Pedro K. Beredjiklian, MD
Purpose The purpose of this study is to quantify finger/digit rotation, overlap, parallelism, and convergence to the scaphoid tuberosity in normal volunteers to establish standards for comparison. Methods We examined 240 uninjured fingers in 30 volunteers. There were 14 men and 16 women with an average age of 35 years. Rotation was determined with the palm flat and fingers extended. Digit overlap, parallelism (angular relationship between the index, ring, and small fingers with respect to the middle finger) and scaphoid convergence were determined with simultaneous flexion of metacarpophalangeal and proximal interphalangeal joints. Linear and angular measurements were performed with imaging software. Results Rotation, parallelism, and scaphoid convergence measurements were similar comparing left with right hands. Rotation: All digits were found to be in supination relative to the horizontal plane; the small fingers averaged 9°, the ring fingers 4°, the middle fingers 8°, and the index fingers 12°. Parallelism: Angular measurements between the middle finger and the small finger averaged 19°, between the middle finger and the ring finger 9°, and between the middle finger and the index finger 11°. Scaphoid convergence: In no hand did all 4 fingers converge onto scaphoid tuberosity. Whereas small and ring finger trajectories averaged ⫺0.1 to 1.2 mm from the scaphoid tubercle, the middle finger averaged 4.0 to 4.6 mm and the index finger 8.1 to 9.5 mm. Overlap: Ninety percent of individuals demonstrated digit overlap, although none covered more than 50% of the adjacent nail plate. Seventy-seven percent of these were bilateral, and 73% were asymmetric. Conclusions This study establishes the normal parameters for digit rotation, overlap, parallelism, and scaphoid convergence. For digit rotation, parallelism, and scaphoid convergence, the contralateral (uninjured) hand can be used reliably for comparison. However, for digit overlap, the contralateral hand should not be used for comparison because of side-to-side asymmetry and variability. (J Hand Surg 2008;33A:873–878. Copyright © 2008 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Diagnostic IV. Key words Digit alignment, digit overlap, digit rotation, finger fracture.
From the Department of Orthopaedics, Division of Hand, Upper Extremity and Microvascular Surgery, UMDNJ-NJ Medical School, Newark, NJ; and the Department of Orthopaedic Surgery, Penn Hand Center, University of Pennsylvania Health System, Philadelphia, PA. Received for publication November 15, 2007; accepted in revised form February 6, 2008. No benefits in any form have been received or will be received related directly or indirectly to the subject of this article. Corresponding author: Virak Tan, MD, Department of Orthopaedics, Division of Hand, Upper Extremity and Microvascular Surgery, UMDNJ-NJ Medical School, 140 Bergen St. (ACC South, D Level, Orthopaedics), Newark, NJ 07103; e-mail: [email protected]. 0363-5023/08/33A06-0011$34.00/0 doi:10.1016/j.jhsa.2008.02.006
ETACARPAL AND PHALANGEAL fractures are common, representing more than 40% of all hand and forearm fractures.1 Although most of these injuries can be successfully treated nonoperatively, certain fractures are better managed with surgery. It is essential to properly assess fracture pattern and stability, as this will guide fracture management. Undertreatment of an unstable fracture can result in deformity and functional limitation.2–5 One of the more ambiguous parameters to measure in a fractured finger is rotational deformity. Various techniques for mea-
M
© ASSH 䉬 Published by Elsevier, Inc. All rights reserved. 䉬 873
874
DIGITAL ROTATION AND ALIGNMENT IN NORMAL SUBJECTS
FIGURE 2: Special hand positioning device that passively holds the MCP and PIP joints in flexion. FIGURE 1: Photograph of the hand in position 1, with the palm flat on a table with fingers extended. The photograph is centered on the middle finger. The rotation of the digit is established by measuring the angle between the nail plate and the horizontal surface (shown for the middle finger).
surement and acceptable parameters have been described,2,4,6,7 and most clinicians determine malrotation of finger fractures by comparison with the uninjured hand. Assuming that there is minimal side-to-side variability, most clinicians determine malrotation of finger fractures by comparison with the uninjured hand. The purpose of this study was to determine whether the aforementioned assumption is correct by quantifying side-to-side variations in digit rotation, digit overlap, and digit convergence to the scaphoid tuberosity in normal volunteers. MATERIALS AND METHODS We examined 240 fingers (index, middle, ring, and small) in 30 volunteers without prior hand injury. There were 14 men and 16 women with an average age of 35 years (range, 20 – 63 years). Digital photographs of the hands were obtained in 2 positions. Position 1 was with the palm flat on a horizontal surface, full finger extension, wrist in neutral alignment, and the camera centered down the axis of the middle finger (Fig. 1). Position 2 was obtained with the hand placed in a special passive positioning device (Fig. 2) allowing simultaneous flexion of the metacarpophalangeal (MCP) and the proximal interphalangeal (PIP) joints. The wrist was in neutral alignment and the camera centered over the interval between the middle and ring fingers (Fig. 3). Rotation was determined by the angle of the nail plate relative to a horizontal plane with the hand in position.1.2,4,6,7 Digit overlap, parallelism, angulation, and convergence to the scaphoid tuberosity were determined with the hand in position 2. Digit
FIGURE 3: Photograph of a hand in position 2, side view. Digit overlap, angulation, and convergence to the scaphoid tuberosity were determined with the hand in position 2 but the photographs were taken vertically (see Figs. 4, 5).
overlap was defined as any obscuring of the nail plate of one finger by an adjacent finger and was graded in 25% increments of the nail plate covered (Fig. 4). Parallelism was determined with respect to the middle finger and defined as the angle formed between a longitudinal trajectory of the middle finger and each of the other fingers (Fig. 5). Scaphoid tubercle convergence was assessed by marking a 1-cm circle around the scaphoid tubercle volarly and then measuring the horizontal distance between it and a line drawn along the trajectory of each digit (Fig. 5). The determination of scaphoid tubercle position was made by 1 observer, and the variability in determining this landmark was not performed. Angular and linear measurements were made on the digital photographs, and analysis was performed by a
JHS 䉬 Vol A, July–August
DIGITAL ROTATION AND ALIGNMENT IN NORMAL SUBJECTS
875
RESULTS The results for digit rotation, parallelism, and scaphoid convergence are presented in Table 1. Rotation Left side to right side digit rotation for each finger was found to be similar and was in relative supination. Rotation of the small finger averaged 8° on the left versus 10° on the right (p ⫽ .12); the ring finger, 4° versus 4° (p ⫽ .79); the middle finger, 8° versus 9° (p ⫽ .48); and the index finger, 12° versus 12° (p ⫽ .98), left versus right, respectively. FIGURE 4: Photograph showing 25% overlap of the ring finger by the middle finger. The black circle marks the palpable scaphoid tubercle volarly.
Parallelism Parallelism relative to middle finger was similar between the 2 sides. Angular measurements between the middle finger and the small finger averaged 19° on the left and 19° on the right (p ⫽ .93); between the middle finger and the ring finger, 9° versus 9° (p ⫽ .67); and between the middle finger and the index finger,10° versus 11° (p ⫽ .54), left versus right, respectively. Scaphoid tubercle convergence In no hand did all 4 fingers converge onto scaphoid tuberosity. However, there remained symmetry with side-to-side comparison; p value for left versus right in the small finger is .26, for the ring finger .06, for the middle finger .49, and for the ring finger .26. See Table 1 for ranges, standard deviations, and 95% confidence intervals for the data.
FIGURE 5: Scaphoid tubercle convergence was assessed by marking the palpable scaphoid tubercle volarly (black circle) and then measuring the horizontal distance between it and a line drawn along the trajectory of each digit. In this photograph, the middle and ring finger trajectories fall onto the tuberosity, but those of the index and small fingers do not. Parallelism is defined as the angle formed between a longitudinal trajectory of the middle finger and each of the other fingers.
single observer (V.T.) on a Macintosh computer using the public domain NIH Image program (developed at the U.S. National Institutes of Health and available on the Internet at http://rsb.info.nih.gov/ nih-image/). t-tests were performed for rotation, parallelism, and scaphoid tuberosity convergence. Chisquare analysis was used to test significance for the nominal variable in digit overlap. A p value less than .05 was considered significant.
Overlap Ninety percent of individuals had partial overlap of the nail plates of at least 1 finger. In 23 subjects, the overlap occurred in both hands. Comparing left with right hands, there was asymmetry (ie, overlap of different fingers) in 22 of 30 subjects (p ⬍ .001). In all digits that had overlap, the nail plate was covered 50% or less. DISCUSSION Indications for fixation of metacarpal and phalangeal fractures of the hand are based on angular and rotational deformity, bony apposition, comminution, and joint stability. Whereas various measurements may be considered acceptable depending on which bone is fractured and the location of the fracture (head, neck, shaft, base), even a small degree of malrotation (0 to 10°) has been an indication for fixation.2,4,6,8 One degree of rotation at the fracture site may translate to 5° at the fingertip,9 and as little as 5° of fracture malrotation can cause 1.5 cm of digital overlap when the fingers are flexed.4 Despite the importance of proper rotational
JHS 䉬 Vol A, July–August
876
TABLE 1. Normal Ranges of Digit Rotation, Parallelism, and Scaphoid Convergence Left Hand
Right Hand
SF
RF
MF
8
4
8
IF
IF
MF
RF
9
4
SF
Digit rotation* (°) Range SD JHS 䉬 Vol A, July–August
95% confidence interval
12
12
10
⫺3 to 22
⫺8 to 17
1 to 15
1 to 25
⫺2 to 25
⫺1 to 24
⫺11 to 16
0.2 to 30
5.5
5.6
4.5
5.5
6.3
6.9
6.9
7.3
5.8 to 9.9
1.7 to 5.8
6.1 to 9.4
9.9 to 14.0
9.6 to 14.4
6.2 to 11.3
1.6 to 6.8
7.7 to 13.1
p value
.12
.79
.48
.98
Digit parallelism†(°) Mean Range SD 95% confidence interval
19
9
N/A
10
11
9
19
10 to 31
2 to 20
N/A
3 to 20
N/A
0 to 17
7 to 39
5.6
4.2
N/A
4.2
4.0
N/A
4.0
6.8
17.3 to 21.2
7.6 to 10.6
N/A
8.9 to 11.9
9.5 to 12.4
N/A
7.2 to 10.0
16.7 to 21.6
—
.54 9.5
4.6
1.2
0.5
p value
.93
.67
0.3 to 18
N/A
Scaphoid convergence‡ (mm) Mean
0
⫺0.1
4.0
Range
⫺10.4 to 6.9
⫺6.2 to 5.2
⫺3.1 to 14.9
2.0
4.5
SD 95% confidence interval p value
2.8 ⫺2.5 to 2.0 .26
⫺0.5 to 3.0 .06
5.7 to 10.1 .49
8.1 0 to 22.0
0 to 21.3
0 to 17.0
⫺0.7 to 6.3
5.6
6.1
4.5
1.9
10.6 to 15.0
12.3 to 16.7
6.8 to 10.8
1.7 to 5.1
.26
SF, small finger; RF, ring finger; MF, middle finger; IF, index finger; N/A, not applicable. *With MCP and PIP joints in extension; rotation relative to horizontal plane; positive value ⫽ supination, and negative value ⫽ pronation. †With finger flexion in device; relative to axis of long finger. ‡With finger flexion in device; positive value ⫽ radial, and negative value ⫽ ulnar to scaphoid tuberosity.
⫺6.3 to 8.5 3.2 ⫺1.0 to 3.5
DIGITAL ROTATION AND ALIGNMENT IN NORMAL SUBJECTS
Mean
DIGITAL ROTATION AND ALIGNMENT IN NORMAL SUBJECTS
alignment, it has been relatively difficult to assess reproducibly.10 Several methods have been used. With MCP and PIP joint flexion, increased finger overlap or scissoring (compared with the contralateral side) indicates malrotation.4 Furthermore, it has been reported that in this position, the fingertips normally point to the scaphoid tuberosity.2,4 Soft tissue swelling alone can cause apparent malrotation. Smith et al.7 simulated soft tissue swelling by injecting saline into the fourth web spaces of noninjured volunteers and assessed digital rotation radiographically and clinically in flexion and extension. After injection, all subjects demonstrated a new onset of digital overlap of the flexed small finger on the ring finger (clinically and radiographically). However, none demonstrated malrotation in extension, and the angles between the nail plates and the horizontal plane remained stable. They concluded that small finger internal rotation/supination (pseudorotation) can be associated with fourth web space edema and could cause falsely increased scissoring/overlap in flexion. In our study, 27 individuals had overlap of the fingers. Although 23 of these were bilateral, 22 were asymmetric in side-to-side comparison. This suggests that comparing the injured side to the uninjured side may not be reliable when assessing digit overlap (scissoring) with finger flexion. Based on our findings, we propose that 50% or less overlap of the nail plate in any digit is within normal limits and does not cause functional difficulties (“50% rule”). It has been written that all digits should point to the scaphoid tubercle when in flexion (ie, scaphoid convergence test); however, there is a discrepancy in how this test should be performed. Some published reports illustrate the maneuver with all fingers flexed simultaneously,2,4 while others had the digits flexed individually.11,12 We performed the test with all the fingers simultaneously flexed, and none of the patients demonstrated convergence of all fingertips to the scaphoid tubercle (Fig. 5). As such the concept of “scaphoid convergence” may only hold true when the digits are flexed individually. We did not identify any statistical differences from side to side for each finger when evaluating scaphoid tubercle convergence or parallelism. Because of this side-to-side symmetry, we believe that comparison of convergence and parallelism to the contralateral hand is a more accurate determination of rotation in those cases where the MCPs and PIPs can be simultaneously flexed. In addition, parallelism measurements can be
877
done independent of the location of the scaphoid tubercle. Another method for assessing digit rotation is measuring the angle of the nail plate relative to the horizontal plane, with the palm flat and fingers extended.2,4,6,7,13 This method obviates the need to try to flex all the fingers down. Our data challenge the notion that digit rotation within 10° of neutral is within normal limits.6,14 We found that all digits were in relative supination with the index finger having an average of 12°, and the range for the small finger was up to 30°. Based on our data, normal digit rotation for the small fingers is 9° ⫾ 6, the ring fingers 4° ⫾ 6, the middle fingers 8° ⫾ 6, and the index fingers 12° ⫾ 6. In our subjects, the average rotation for each digit is within 5° of the findings of Bansal and Craigen.13 The current study establishes the normal ranges of rotation for each digit and side-to-side similarity, which can aid in proper assessment of rotation (Table 1). In a similar study, Bansal and Craigen13 describe the normal parameters of digital rotation based on the assessment of fingernail orientation with the fingers in full extension and with the MCP joints fully flexed. These authors conclude that the orientation of the fingernails can vary within and between individuals. In our study, digit alignment was also measured with the digits fully flexed, in a manner that mimics the clinical setting more accurately and addresses the main concern with malrotation, which is digital overlap.15 We demonstrate that although in most circumstances comparison with the contralateral (uninjured) side may be sufficient, digit overlap is often asymmetric. Side-to-side comparison in this case may be misleading, but clinicians can use the “50% rule” as a guide. Furthermore, in the instance of a previous or acute contralateral hand injury, knowledge of normal ranges can be very helpful. With critical evaluation of normal anatomic alignment, our data challenge the widely accepted teaching that all fingertips point to the scaphoid tuberosity on attempted fist formation. A limitation of our study is that we did not have a comparison group with fractures. However, the purpose of the study was to establish normal values to be the baseline when assessing finger alignment. A second limitation is that the reference landmarks were determined by palpation (of the scaphoid tuberosity) and visual inspection, which could have been subjected to bias or errors. However, we believed that this represented the real-life clinical situation when trying to assess finger rotation and alignment. An additional limitation is the lack of reliability coefficients for
JHS 䉬 Vol A, July–August
878
DIGITAL ROTATION AND ALIGNMENT IN NORMAL SUBJECTS
the data because we only had a single observer. Finally, in the absence of any clinical data, we are unable to determine what degree of digital malrotation is clinically important. REFERENCES 1. Henry MH. Fractures and dislocations of the hand. In: Bucholz RW, ed. Rockwood & Green’s fractures in adults. 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2006:772. 2. Ashkenaze DM, Ruby LK. Metacarpal fractures and dislocations. Orthop Clin North Am 1992;23:19 –33. 3. Diao E, Welborn JH. Extraarticular fractures of the metacarpals. In: Berger R, Weiss APC, eds. Hand surgery. 1st ed. Philadelphia: Lippincott Williams & Wilkins, 2004:139 –151. 4. Jupiter JB, Axelrod TS, Belsky MR. Fractures and dislocations of the hand. In: Browner BB, ed. Skeletal trauma. 3rd ed. Philadelphia: Saunders, 2003:1153. 5. Stern PJ. Fractures of the metacarpals and phalanges. In: Green D, Hotchkiss R, Pederson W, Wolfe S, eds. Green’s operative hand surgery. 5th ed. Philadelphia: Elsevier, 2005:277–341.
6. Royle SG. Rotational deformity following metacarpal fracture. J Hand Surg 1990;15B:124 –125. 7. Smith NC, Moncrieff NJ, Hartnell N, Ashwell J. Pseudorotation of the little finger metacarpal. J Hand Surg 2003;28B:395–398. 8. Barton NJ. Fractures of the hand. J Bone Joint Surg 1984;66B:159 – 167. 9. Opgrande JD, Westphal SA. Fractures of the hand. Orthop Clin North Am 1983;14:779 –792. 10. Bain GI, Guy SD. Derotation taping for metacarpal and phalangeal fractures. Tech Hand Up Extrem Surg 2003;7:130 –132. 11. Smith P. Lister’s the hand: diagnosis and indications, 4th ed. New York: Churchill Livingstone, 2002: 13. 12. Chang J. Plastic surgery: contributions to hand surgery. In: Hentz V, ed. Mathes’ plastic surgery, volume VII: the hand and upper limb, part 1, 2nd ed. Philadelphia: Elsevier, 2006:20. 13. Bansal R, Craigen MAC. Rotational alignment of the finger nails in a normal population. J Hand Surg [Eur] 2007;32:80 – 84. 14. Smith RJ and Peimer CA. Injuries to the metacarpal bones and joints. Adv Surg 1977;11:341–374. 15. Rosenwasser MP, Quitkin HM. Malunion and other posttraumatic complications in the hand. In: Berger R, Weiss APC, eds. Hand surgery, 1st ed. Philadelphia: Lippincott Williams & Wilkins, 2004: 207–231.
JHS 䉬 Vol A, July–August