Forearm rotation is a function of the complex interaction among the radius, ulna, and interosseous membrane. During supination and pronation, the radius rotates around a longitudinal axis of the forearm that passes through the center of the radial head and the posterolateral aspect of the ulnar head . These mechanics allow for nearly 180° of rotation, but fractures of both bones of the forearm can be detrimental if they result in an aberration of this anatomic relationship. Matthews et al.  demonstrated that a 10° angulation deformity in one or both of the forearm bones results in a 20° loss of pronation or supination. Furthermore, Abe et al.  found that an extension deformity of the radius results in pronation restriction due to osseous impingement, while a valgus deformity of the ulna results in supination restriction due to tightness of the central band of the interosseous membrane. In addition, rotational deformity of more than 30° also has been shown to impede pronosupination . Loss of forearm rotation can impede upper extremity function and activities of daily living.
Determining the degree of forearm bowing aids the surgical treatment of forearm fractures. The ulna translates approximately 9° in the varus or valgus plane at the elbow, allowing the ulnar head to accommodate the rotating radius distally . Furthermore, to allow for rotation around the ulna, the radius has a bowed shape, with a mean maximal bow of 15.3 ± 0.3 mm . Restoration of the radial bow is a well-documented goal of radius fixation [5, 6, 13, 15], and the surgeon frequently selects a pre-contoured plate or bends the radial shaft plate in the coronal plane to accommodate radial bow. Although studies have described the ulnar anatomy as irregular with complex angulation at the ends [2, 3, 16], the ulna is generally thought of as a “straight bone” . Yet, anecdotally, the surgeon must bend the ulnar shaft plate at times to correctly fit the ulna, indicating that an aspect of ulnar bow is present. If this ulnar bow is not addressed, a mal-reduction of the ulna will cause functional limitations as previously stated.
Schemitsch and Richards  described a method to determine radial bow on AP radiographs of the forearm, and Lincoln and Mubarak  described a method to determine ulnar bow on lateral radiographs of the forearm in children. To our knowledge, there have been no studies on adults to demonstrate the amount of ulnar bow in normal forearms. Having this information would allow for better preoperative planning, which may lead to enhanced anatomic fixation of ulnar fractures. Along with information regarding the curvature of the radius, these data may allow for improved functionality after fixation of forearm fractures overall.
Therefore, we asked: (1) To what degree is the ulna bowed in the coronal and sagittal planes in normal adult forearms? (2) To what degree is the radius bowed in the coronal plane in normal adult forearms?
Patients and Methods
All radiographs of the forearms of adults (range 18 to 99 years) taken at a Level I tertiary care center between January 1, 2017 and December 31, 2017 were obtained. Radiographs were excluded if evidence of a fracture or post-fracture fixation was found, or if a patient had missing AP or lateral images. We chose radiographs that were recorded using the best radiographic technique. In this analysis, a “suboptimal” radiographic technique referred to either a rotated view or inappropriate forearm positioning (Fig. 1). Each radiograph in our final analysis, therefore, was acquired using the optimal technique and demonstrated normal forearm anatomy in the supinated and extended positions in the coronal and sagittal planes (Fig. 2A-B). On the AP radiograph, the radial styloid and biceps tuberosity should be 180°, and the trochlea and capitulum were viewed in profile. On the lateral radiograph, the distal radius and ulna were superimposed, and the trochlea and capitulum were superimposed. Radiographs were included only after all four authors independently inspected and unanimously approved them.
We reviewed 974 radiographs in our initial search. In all, 340 radiographs demonstrated evidence of fracture, and 142 demonstrated surgically fixed fractures, which we excluded. We excluded an additional 394 patients because of suboptimal positioning or missing AP or lateral images, resulting in 98 patients who were included in our final analysis. The mean age of our patients was 38 ± 15 years. There were approximately equal numbers of male and female patients (50 males and 48 females) and radiographs of right and left forearms (53 and 45, respectively).
Our protocol for measuring coronal ulnar bow was adapted from Schemitsch and Richard’s  assessment of radial bow. It is best measured on standard AP radiographs of the forearm in full supination. First, we drew a line (UAy) from the most-radial aspect of the proximal ulna to the most-radial aspect of the ulnar styloid (Fig. 3A-B). Then, we drew a perpendicular line (UAr) at the point of the maximal ulnar bow to UAy. A third line (UAx) was drawn from a line perpendicular to the previously recorded, most-radial aspect of the proximal ulna. Each line was measured in millimeters. The maximal coronal ulnar bow was recorded as a percentage of the ulnar length using the formula UAr/UAy x 100. The location of the maximal coronal ulnar bow was calculated as UAx/UAy x 100.
Our protocol for measuring sagittal ulnar bow was adapted from that described by Lincoln and Mubarak . Bowing in this plane is best measured on standard lateral radiographs of the forearm in neutral rotation. First, we drew a line (ULy) along the dorsal border of the ulna at the level of the olecranon to the ulna styloid (Fig. 4A-B). We drew a perpendicular line (ULr) at the point of maximal ulnar bow to ULy. The point of maximal bow was determined by assessing the point at which the inferior border of the ulna is the farthest from the ULy line. A third line (ULx) was then drawn from the perpendicular line to the previously recorded point on the olecranon. Each line is measured in millimeters. The maximal sagittal ulnar bow was recorded as a percentage of the ulnar length, measured proximally to distally, using the formula ULr/ULy x 100. In addition, the contour of the ulna was described as either convex or concave with apex directed ulnarly from the straight ulnar line.
Radial bow was assessed using the methods of Schemitsch and Richards . In the coronal plane, we first drew a line (Ry) from the most distal and ulnar aspect of the radius at the wrist to the bicipital tuberosity. We drew a perpendicular line (Rr) from the point of maximal radial bow to this line. The point of maximal radial bow was established by assessing the point at which the medial border of the radius is farthest from the Ry line. The location of maximal bow was then measured as the distance from the bicipital tuberosity to the previously measured perpendicular line (Rx). It was recorded as a percentage of the length of the entire radius, measured proximally to distally.
The four study authors independently performed all measurements. The authors varied in expertise; two were orthopaedic hand fellowship-trained attendings, one was a senior orthopaedic resident, and one was a medical student. Interobserver reliability was calculated for radial and ulnar bow measurements and reported as the interclass correlation coefficient (ICC). Our analysis revealed positive, systematic agreement among all researchers for ulnar bow in the coronal and sagittal planes (ICC = 0.96 and 0.97, respectively) and for radial bow (ICC = 0.90). These results fall under the qualitative definition of “excellent” interobserver reliability .
We conducted an independent sample t-test to compare the mean radial bow measurements from our dataset with those reported by Schemitsch and Richards . Data were analyzed using the Statistical Package for the Social Science (IBM Corp, Armonk, NY, USA).
Ulnar Bow in Uninjured Adult Forearms
The mean maximal coronal ulnar bow was 7 ± 2 mm in the ulnar direction. The mean maximal coronal ulnar bow was 3.0 ± 0.9% of the total ulnar length. The location of maximal coronal bow was 75 ± 8% of the total length of the ulna, measured proximally to distally (Table 1). The mean maximal sagittal ulnar bow was 6 ± 3 mm. The maximal sagittal ulnar bow was 3 ± 3% of the total ulnar length. The location of the maximal sagittal bow was 39 ± 13% of the total length of the ulna, measured proximally to distally (Table 1). Of 98 radiographs, one observer (DVC) identified two concave-shaped ulnas in the sagittal plane, two (MMV, JH) identified three, and one (IHA) identified four. The remaining ulnas were convex shaped.
Bowing of the Radius in Uninjured Adult Forearms
There was mean radial bow of 14 ± 2 mm in the coronal plane. The maximal radial bow was 59 ± 5% of the total length of the radius (range 16 to 91) measured proximally to distally (Table 2).
Forearm rotation is a function of the complex interaction between the radius and the ulna that allows for nearly 180° of rotation. During supination and pronation, the radius rotates around the ulna . Prior studies have assessed ulnar morphology in pediatric patients [8, 10], and radial bow in adults is also well recognized . The ulna is considered to be a “straight bone;” however, it is bowed in the coronal and sagittal planes throughout its length. A failure to recreate the ulnar bow during fracture fixation can cause a malunion, which has been shown to result in a loss of rotation and thus impede upper extremity function [5, 6, 12, 13, 15]. Knowledge of the ulnar bow would allow for better preoperative planning, which may lead to greater anatomic fixation of ulnar fractures.
There are several important limitations to our study. Obtaining acceptable AP and lateral forearm radiographs was difficult, as 80% of reviewed radiographs were deemed unacceptable. Although we attempted to include radiographs oriented perfectly in the standard AP and lateral positions, we needed to establish an acceptable level of rotation and angulation, which may have slightly affected our results. Changes in rotation of the bones or angulation of the x-ray beam will add error to the measurements. This is due to structures being more or less prominent in supinated, pronated, or neutral positions. However, we do believe that our sample size will limit gross differences in the true measurements. In addition, radiographs only provide data in 2-D; thus, 3-D bowing of the ulna could not be studied accurately. Use of CT or MRI could better elucidate these details. Despite these limitations, plain radiographs for assessment of forearm bowing were used in the previous studies from which we adopted our measurement technique [10, 15]. In addition, 2-D radiographs are usually only available in the operating room when the surgeon is judging the fracture reduction and is thus more applicable.
Another limitation is that we chose to examine adult patients who initially presented to our hospital. Most of our patients are African American, Hispanic and Caucasian, which may skew our results. There may be a racial variation in forearm bone bowing, similar to what has been shown in femoral bowing in the Asian population . Thus, the results of our study may not be generalizable to patients who were not included. However, due to the large size, varying ages, and similar ratios of sexes of our sample population, other potential variations are likely excluded.
The ulna is bowed in the coronal and sagittal planes. Coronal bow is clinically more relevant than sagittal bow because the plate is typically placed in the dorsal or volar plane. Because no previous studies have evaluated these data, we could not compare our results with those of other studies. Our findings of sagittal bow in adults contrasts with that of pediatric patients, which has been reported to be 1.07 ± 0.7 mm  and no greater than 3.86 mm . The degree of sagittal ulnar bow is thus more pronounced in adults than in the pediatric population. In addition, most adult patients had a convex-shaped ulna. Collectively, these data suggest that a sagittal bow of the ulna changes during a patient’s lifetime, leading to more pronounced bow in adulthood. This decreasing variability may be owing to growth and muscular activity that exert pulling forces on developing bone throughout a patient’s lifetime.
Radial bowing is well described in the previous evidence . The mean radial bow in the coronal plane in our cohort differs from that reported previously (14 mm versus 15 mm) when measured using the same procedure. These differences may not be clinically relevant. Despite these differences, there is a clear need to address the contour of the radius because fixation with a straight plate may lead to poorer anatomic reduction and diminished function.
Historically, emphasis has been placed on anatomic restoration of the radius to achieve baseline ROM, and some studies provided evidence to support this practice [6, 12]. Other studies, however, suggest that the combined effects of radial and ulnar mal-alignment may have a greater clinical impact than mal-alignment of either bone alone [7, 14, 18]. Several studies have found that the amount of lost forearm movement is directly proportional to the loss of normal alignment [14, 15, 18]. Small mal-alignments may have large effects on rotation, with reports showing that as little as 20° of angulation in any direction after fractures resulted in a clinically substantial loss of forearm rotation . Loss of supination and pronation results in worse upper extremity function and may impede activities of daily living and reduce patient satisfaction [6, 13, 15].
The current study suggests that the ulna is bowed in the coronal and sagittal planes throughout its length and has a convexity directed dorsally in the sagittal plane. This knowledge should be considered during fracture fixation to restore normal forearm anatomy. Restoring normal anatomy can be accomplished using a pre-contoured plate or by applying a bend to the plate to match the ulnar bow. This study also demonstrates that the location of the ulnar fracture should be considered since the maximal ulnar bow was found to be always distal to the maximal radial bow. Restoration of the native ulnar bow will prevent malunion and thus maintain forearm rotation and limit functional detriments.
The ulna is not a “straight bone,” as is commonly thought, but rather has a bow in both the coronal and sagittal planes. Knowledge of the standard ulnar bow may be pivotal to prevent ulnar malunion during surgery. Future research using these data in preoperative planning may lead to changes in plate contouring and clinical outcomes in forearm fracture management.
1. Abe S, Murase T, Oka K, Shigi A, Tanaka H, Yoshikawa H. In vivo three-dimensional analysis of malunited forearm diaphyseal fractures with forearm rotational restriction. J Bone Joint Surg Am. 2018;100:e113.
2. Beşer CG, Demiryürek D, Özsoy H, Erçakmak B, Hayran M, Kızılay O, Özsoy A. Redefining the proximal ulna anatomy. Surg Radiol Anat. 2014;36:1023-1031.
3. Chapleau J, Balg F, Harvey EJ, Ménard J, Vauclair F, Laflamme GY, Hebert-Davies J, Rouleau DM. Impact of olecranon fracture malunion: Study on the importance of PUDA (proximal ulna dorsal angulation). Injury. 2016;47:2520-2524.
4. Cicchetti DV. Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychol Assess. 1994;6:284-290.
5. Freese KP, Faulk LW, Palmer C, Baschal RM, Sibbel SE. A comparison of fixation methods in adolescent patients with diaphyseal forearm fractures. Injury. 2018;49:2053-2057.
6. Goldfarb CA, Ricci WM, Tull F, Ray D, Borrelli J Jr. Functional outcome after fracture of both bones of the forearm. J Bone Joint Surg Br. 2005;87:374-379.
7. Grace TG, Eversmann WW Jr. Forearm fractures: treatment by rigid fixation with early motion. J Bone Joint Surg Am. 1980;62:433-438.
8. Jeong WK, Lee DH, Kyung BS, Lee SH. Factors affecting assessment of ulnar bowing in radiography. J Pediatr Orthop. 2012;32:48-53.
9. Kasten P, Krefft M, Hesselbach J, Weinberg AM. How does torsional deformity of the radial shaft influence the rotation of the forearm? A biomechanical study. J Orthop Trauma. 2003;17:57-60.
10. Lincoln TL, Mubarak SJ. "Isolated" traumatic radial-head dislocation. J Pediatr Orthop. 1994;14:454-457.
11. Matsuki KO, Matsuki K, Mu S, Sasho T, Nakagawa K, Ochiai N, Takahashi K, Banks SA. In vivo 3D kinematics of normal forearms: analysis of dynamic forearm rotation. Clin Biomech (Bristol, Avon). 2010;25:979-983.
12. Matthews LS, Kaufer H, Garver DF, Sonstegard DA. The effect on supination-pronation of angular malalignment of fractures of both bones of the forearm. J Bone Joint Surg Am. 1982;64:14-17.
13. Sarmiento A, Ebramzadeh E, Brys D, Tarr R. Angular deformities and forearm function. J Orthop Res. 1992;10:121-133.
14. Schemitsch EH, Jones D, Henley MB, Tencer AF. A comparison of malreduction after plate and intramedullary nail fixation of forearm fractures. J Orthop Trauma. 1995;9:8-16.
15. Schemitsch EH, Richards RR. The effect of malunion on functional outcome after plate fixation of fractures of both bones of the forearm in adults. J Bone Joint Surg Am. 1992;74:1068-1078.
16. Shin WJ, Kim JP, Yang HM, Lee EY, Go JH, Heo K. Topographical anatomy of the distal ulna attachment of the radioulnar ligament. J Hand Surg Am. 2017;42:517-524.
17. Slutsky DJ. Operative Orthopedics of the Upper Extremity.
New York, NY: McGraw-Hill Education; 2014. 453-462.
18. Wilson FC, Dirschl DR, Bynum DK. Fractures of the radius and ulna in adults: an analysis of factors affecting outcome. Iowa Orthop J. 1997;17:14-19.
19. Yau WP, Chiu KY, Tang WM, Ng TP. Coronal bowing of the femur and tibia in Chinese: its incidence and effects on total knee arthroplasty planning. J Orthop Surg (Hong Kong). 2007;15:32-36.