Journal of Orthopaedic Research 8:167-174 Raven Press, Ltd., New York 0 1990 Orthopaedic Research Society

Excursion of the Flexor Digitorum Profundus Tendon: A Kinematic Study of the Human and Canine Digits S. Horibe, S. L-Y. Woo, J. J. Spiegelman, J. P. Marcin, and R. H. Gelberman Orthopaedic Bioengineering Laboratory, University of California, San Diego and Malcolm and Dorothy Courts Institute for Joint Reconstruction and Research, La Jolla, California, U.S.A.

Summary: The most common problem following primary flexor tendon repair is the failure of the tendon apparatus to glide, secondary to the formation of adhesions. Early motion following tendon repair has been shown to be effective in reducing adhesions between the tendon and the surrounding sheath. Therefore, it is important to determine the amount of flexor tendon excursion along the digit during joint motion. In this study, the excursion between the flexor digitorum profundus (FDP) tendon and the sheath was examined in both human and canine digits. Based on roentgenographic measurements and joint kinematic analysis, the motion of the bones, the FDP tendon, and the sheath were measured with respect to joint rotations. It was found that the canine flexor tendon apparatus behaved similarly to that of the human for the motions studied. The amount of tendon excursion was very small in regions distal to the joint in motion (approximately 0.1 mm/lO" of joint rotation). There was little displacement of the sheath (0.2-0.3 mm), except at the metacarpal joint region during metacarpophalangeal (MCP) joint motion and at the proximal interphalangeal (PIP) joint region during PIP joint motion. Tendon excursion relative to the tendon sheath was the largest in zone I1 during PIP joint rotation (1.7 mm/lO" of joint rotation). These results suggest that PIP joint motion may be most effective in reducing adhesions following tendon repair in zone 11. Key Words: Tendon-Kinematics-Biomechanics-Excursion-Flexor-Sheath.

rehabilitation programs, interest has been renewed in the repair of the flexor tendon. Complications, however, still remain common. One of the problems following primary flexor tendon repair is the formation of adhesions. The adhesions between the flexor tendon and the surrounding sheath restrict the normal gliding between these two tissues and lead to dysfunction of the digit. The roles of various factors such as suturing technique, suture material (12), and postoperative rehabilitation regimens have been investigated in order to reduce such adhesions. Of these, postoperative passive mobilization has been the most extensively investigated, both clinically (6,14,20,21) and experimentally (7,8,10,11,15,22). Laboratory studies indi-

Historically, results of primary flexor tendon repair have not been satisfactory in zone 11, defined as the region from the distal palmar crease to the insertion of the flexor digitorum superficialis (FDS) tendon at the middle phalanx (4,13,20,21). Thus, tendon graft has been the recommended procedure rather than primary tendon repair. However, with recent advances in surgical technique and improved

Received December 16, 1988; accepted April 26, 1989. Address correspondence and reprint requests to Dr. S. Woo, University of California, San Diego, Division of Orthopaedics and Rehabilitation, M-030, La Jolla, CA 92093, U.S.A. The present address of Dr. R. H. Gelberman is Massachusetts General Hospital, Harvard Medical School, Boston, MA.

167

S.HORIBE ET AL.

168

cate that protected passive motion of the healing tendon not only helps to decrease the amount of adhesions between the tendon and the surrounding sheath, but also improves flexor tendon healing by increasing its strength. However, quantitative information regarding a treatment to achieve optimal results during the rehabilitation period is not yet available. For example, the magnitude of tendon excursion for a given amount of joint flexion during rehabilitative passive motion remains unknown. The question as to which joint to flex and how much flexion is required to attain the desired amount of excursion has been primarily addressed only in qualitative terms. In order to gain quantitative data on the displacements of the tendon and sheath, we designed this study (a) to determine tendon excursion relative to sheath along the digit, based on experimental measurement and kinematic analysis; (b) to determine the joint motion that would most effectively maximize tendon excursion relative to sheath; and (c) to compare data for human digits to those of the canine in order to assess the similarities and/or differences between them. MATERIALS AND METHODS Specimen Preparation For the human study, the index and middle fingers from six human cadaver hands were used. For the canine study, the third digit of 12 forepaws was used. The entire digit (or forepaw) together with muscle and skin attachments were wrapped in saline-soaked gauze, sealed in plastic bags, and frozen at - 20°C for no more than 1 month. Before testing, the specimen was thawed at 4°C overnight. The digit was disarticulated at the metacarpal joint, and the flexor digitorum profundus (FDP) tendon was transected at the level of the carpal tunnel. The flexor digitorum superficialis (FDS) tendon was transected midway along the metacarpal bone, and the extensor tendon was cut at the proximal end of the metacarpal bone. All soft tissues proximal to the distal interphalangeal (DIP) joint were removed except for the two flexor tendons, one extensor tendon, and the accompanying ligaments, tendon sheath, pulleys, and joint capsules. For roentgenographic images, the metacarpal and the three phalangeal bones, the FDP tendon, and the tendon sheath were labeled with radiopaque markers (Fig. 1). Three 0.7 mm diameter Kirschner

J Orthop Res, Vol. 8, No. 2,1990

Middle phalanx Metacarpal bone

Proximal phalanx

Distal Dhalanx

Sheath

FIG. 1. Typical location for radiopaque markers placed in the bones, FDP tendon and tendon sheath of a human middle finger. 0 Bone marker; -tendon marker; 9 sheath marker.

wire markers were imbedded in each of the four bones. To label the tendon, a 2-3 mm section of suture (3-0 stainless steel wire) was inserted through the sheath into the tendon using a 23 gauge needle. Nine such markers were implanted along the FDP tendon from the metacarpal area to the DIP joint. The locations of the tendon markers were not known until the roentgenogram was taken. In addition, seven markers (5-0 stainless steel wire with their needles) were sutured to the tendon sheath along its entire length, with no specific reference to the pulleys. Specimen Mounting The specimen was mounted on a specially designed stand, as shown in Fig. 2. In this study, we assumed that the digital structure consisied of a system of four links with three coplanar joints. The specimen was carefully mounted so that the digit was as parallel to the surface of the stand as possible during motion. A grid that was built into the device was used to calibrate the actual displacements of the markers. A preliminary study showed that this method resulted in a marker resolution of 0.01 mm (19). Sutures were attached to the FDP tendon and the extensor tendon. A constant load of 2 N then was applied along the FDP tendon, which was allowed to creep for 30 min before measurements were made. Care was taken to insure that all soft tissues were kept moist with saline gauze at all times. Roentgenograms Four series of two roentgenograms each were taken using a Faxitron (Model 43805N, Hewlett Packard Co., McMinnville, OR, U.S.A.) to study tendon and sheath displacements. In the first series, all three joints, i.e., metacarpophalangeal (MCP), proximal interphalangeal (PIP) and distal interphalangeal (DIP) joints were allowed to rotate freely. In

FLEXOR TENDON EXCURSION

169

FIG. 2. Schematic diagram of the stand used for experiments. A constant load of 2 N was applied along the FDP tendon by the weight.

,

the remaining three series, only one of the three joints was left free to rotate while the other two joints were fixed at approximately 90" of flexion with 0.9 mm Kirschner wires. The first roentgenogram in each series was taken with the suture to the extensor tendon pulled and fixed to the device such that the digit was in full extension. For the second roentgenogram, the tension to the extensor tendon was removed, allowing the 2 N load on the FDP tendon to pull the digit into flexion. Digitization and Calculation A photographic slide of each roentgenogram was made and projected onto a digitizing board. The resolution of the digitizer was 0.01 mm (19) and the error of digitizing the markers on the specimens was within 0.1 mm. First, the calibration grid was digitized to perform an affine transformation on the position values of the projected markers. The markers of the four bones, the FDP tendon, and the sheath were then digitized. The plane of joint flexion was assumed to be parallel to the surface of the device (two-dimensional analysis). The skeletal structure was assumed to be a multilink system with three coplanar revolute joints. The angular rotation (0) and the center of rotation (COR)were calculated for all three joints using a rigid body method (18). The data obtained for the initial (extension) and final (flexion) positions of all of the markers were stored in a computer. Initially, to synchronize the reference coordinate systems, the data file from the flexion position was translated and rotated such that the positions of the metacarpal bone markers were equivalent for the flexion and extension data files. The 0 and the position of the COR of the joint could then be calculated by knowing the position of two markers on the rigid body before and after its rotation (18). Using two of the three markers em-

u bedded in each bone, the 0 and the COR of each joint were calculated from the two data files. The third marker in each bone was used to insure the accuracy of the calculated COR with follow-up calculations. For the regions proximal and distal to the joint in motion, different calculations must be used to compute displacements. If the tendon marker moves in the region proximal to the joint in motion (point A in Fig. 3a), the change in position of the tendon a.

L

b.

c

FIG. 3. (a) Tendon excursion in the region proximal to a specific joint in motion. A is the tendon marker in extension, while A' is the same tendon marker in flexion. (b) Tendon excursion in the region distal to a specific joint in motion. B is the tendon marker in extension. 6' is the same tendon marker in flexion. B is the tendon marker rotated about the COR.

J Orthop Res, Vol. 8, No. 2 , 1990

S. HORIBE ET AL.

170

marker (from A to A') is used to describe the tendon excursion. Note that A is the position of the tendon marker within the digit in extension and A' is the position of the tendon marker in flexion. When the displacement of the tendon marker is in the region distal to the joint in motion (point B in Fig. 3b), its motion is a combination of rigid body rotation and linear translation. For example, B is the tendon marker within the digit in full extension, B' is the same tendon marker with simple rotation about the COR, and B" is the tendon marker after it has been rotated and translated following flexion of the digit. Therefore, the actual tendon excursion is represented by the distance between B' and B". Knowing the positions of markers for the tendon, sheath, and bones, the displacements of the tendon and sheath were obtained in a similar manner. Specifically, the amount of tendon excursion, sheath displacement, and tendon excursion relative to the sheath at five regions along the digital structure were determined (Fig. 4). The five regions were (a) region C: the metacarpal area, defined by the markers proximal to the MCP joint; (b) region M: the MCP joint area defined by the markers around the MCP joint area; (c) region PP: the proximal phalanx area defined by the markers between the MCP and PIPjoints; (d) region P: the PIP joint area defined by the markers around the PIP joint; and (e) region MP: the middle phalanx area defined by the markers distal to the PIP joint. For statistical analysis, a one-way analysis of variance (ANOVA) was performed to compare the tendon excursion and tendon excursion relative to the sheath in different regions. Because of the limited range of motion studied, a linear relationship between excursion and the angular rotation was assumed, i.e., E = KO, where E is the tendon excursion, 8 is the angular rotation, and K is a constant. Although the linear relationship Middle phalanx Metacarpal bane

Proximal phalanx

Distal phalanx

between tendon excursion and angular rotation has not been demonstrated in this study, several previous studies supported this finding (1,2,4). Therefore, using the superposition principle, a linear equation was derived for the tendon excursion of the three joint system by the summation of those for single joint motion, i.e., Etotal =

Klol +

w

2

(1)

+ K383

where Etotalis the total tendon excursion (the motion of the most proximal tendon marker), and o,, o,, and 8, are angular rotations of the MCP, PIP, and DIP joints, respectively (19). K , represents the constant for MCP joint rotation, K , represents PIP joint rotation, and K3 represents DIP joint rotation. These values were obtained based on the tendon excursion at the particular metacarpal region divided by the angular rotation for the joint. The respective constant coefficients K 1 , K,, and K , were derived from the data for the isolated joint motions. The validity of this equation had been checked previously by Spiegelman et al. (19) by comparing the total excursion calculated from the individual joint motions with the total excursion measured from the digits when all three joints were free to rotate. RESULTS

For each type of isolated joint motion tested in the human digits, the amount of tendon excursion, sheath displacement, tendon excursion relative to the sheath, and the angular rotation of the single joint are summarized in Table 1. With 2N of load, the angular rotation for the MCP, PIP, and DIP joints were 81.0 k 6.4,50.2 k 10.1, and 21.8 2.1", respectively (mean 2 SEM). When the MCP joint was free to rotate, the average tendon excursion in the proximal regions C and M were found to be 17.8 2 1.6 and 8.8 ? 1.0 mm, respectively. There was little excursion distal to the MCPjoint. The value of tendon excursion in region C was greater than those in regions distal to the MCP joint (p < 0.001). When the PIP joint was free to rotate, the amount of tendon excursion was 9.7 ? 1.7 mm for region C, 9.2 1.6 mm for region M, and 9.0 2 1.6 mm for region PP. Again, there was little excursion in the regions distal to the PIP joint. The values of tendon excursion in regions proximal to the PIP joint were greater than those in region PP (p < 0.001). When the DIP joint was free to rotate, the amount of the tendon excursion in each of the regions was small (in the range of 2.2-2.7 mm).

*

*

I

I

C

I

I

. . " w ' + Y -

M

PP

!

!

P MP

FIG. 4. Schematic diagram describing the anatomical regions of the human middle finger. C: metacarpal region, M: metacarpophalangeal joint region, PP: proximal phalanx region, P: proximal interphalangeal joint region, MP: middle phalanx region.

Orthop Res, Vol. 8, No. 2 , 1990

171

FLEXOR TENDON EXCLJRSION

TABLE 1. The amounts of the tendon excursion, the sheath displacement, and the tendon excursion relative to the sheath of human fingers Single joint rotations MCP joint 1. Angular rotation (deg) 2. Tendon excursion (rnrn) Regions C M PP P MP 3. Sheath displacement (rnrn) Regions C M PP P MP 4. Tendon excursion relative to the sheath (rnrn) Regions C M PP P MP

The displacements of the tendon sheath under the same loading conditions were different from those for the tendon. With MCP joint rotation, the sheath displacement was highest in region C (11.5 t 1.4 mm). For the cases of DIP and PIPjoint rotation, the sheath motion was generally small (less than 1 mm). As for the tendon excursion relative to the sheath, the highest values for MCP joint rotation were found in region C (6.5 + 0.8 mm) and region M (6.4 & 0.8 mm). In the case of PIPjoint rotation, these values were even higher (more than 8 mm) in regions C, M, and PP. During DIPjoint rotation, the values for tendon excursion relative to the sheath were small for all five regions. To compare the effectiveness of individual joint motion on tendon excursion, the values were expressed in terms of the angular rotation (Fig. 5). When the MCP joint was free to rotate, tendon excursions per 10" of angular rotation in regions C and M were found to be 2.1 t 0.1 and 1.1 2 0.1 mm, respectively. Those distal to the MCP joints were found to be small. When the PIP joint was free to rotate, the amount of tendon excursion was about 1.9 mm/lOoat the regions proximal to the PIP joint. During motion of the DIP joint, the excursion values were all on the order of 1.1 m d l 0 " . Similarly, tendon excursion relative to the sheath was normalized by the angular rotation (Fig. 6). During MCP joint motion, values in regions C and M were both 0.8 mm/l0". There was little relative excursion in the regions distal to the MCP joint. In

PIP joint

81.0 i 6.4

50.2

?

DIP joint

10.1

21.8

f

2.1

17.8 f 1.6 8.8 1.0 1.3 f 0.3 1.3 f 0.4 1.4 k 0.4

9.7 k 1.6 9.2 2 1.6 9.0 f 1.6 6.0 f 1.4 0.8 t 0.2

2.7 f 0.3 2.4 2 0.3 2.5 ? 0.3 2.2 0.2 2.2 2 0.2

11.5 f 1.4 6.1 i 0.7 1.0 0.2 1.4 f 0.4 1.5 0.4

1.2 f 0.2 0.9 i 0.2 1.3 ? 0.5 3.6 k 0.7 1.2 t 0.2

0.6 0.6 0.6

6.5 f 0.8 6.4 t 0.8 1.0 ? 0.2 0.5 f 0.1 0.4 2 0.1

8.8 f 1.6 8.8 t 1.6 8.1 ? 1.6 4.4 f 0.8 0.9 2 0.1

2.3 i 0.3 2.3 f 0.3 2.3 k 0.3 1.6 ? 0.2 1.8 f 0.3

*

*

*

f 0.1 ? 0.1 4 0.1

0.5 i 0.1

0.7

f 0.2

the case of PIP joint motion, the values obtained for regions C, M, and PP were 1.7 ? 0.1, 1.7 t 0.1, and 1.6 ? 0.1 mm/lOo, respectively. The PIP joint motion resulted in the largest tendon excursion relative to the sheath in zone 11. During DIP joint motion, the values were similar among all five regions (1 .O mm/ 1Oo). When all three joints were free to rotate under the applied load, the angular rotations were 104.5 2 5.2, 32.3 2 7.2, and 15.0 f 4.7" for the MCP, PIP, and DIP joints, respectively. The average displacement of the most proximal tendon marker was 30.4

C

M

PP

P

MP

regions along the digit FIG. 5. Amount of tendon excursion per 10" of joint motion for human fingers. Single joint motion: 0 MCP joint; El PIP joint, DIP joint.

J Orthop Res, Vol. 8, No. 2 , 1990

172

S . HORIBE ET AL.

eral, the values for the canine digit were significantly lower than those for the human digit, but the trends for the angular rotation and tendon excursion were identical for the two species. As in the human case, the tendon excursion was found to be minimal distal to the joint in motion. When plotted as excursion per 10" of joint rotation, the amount of tendon excursion relative to the sheath in zone I1 was the highest during PIP joint motion (Fig. 8). This finding again agreed with the results obtained for human digits. C

M

PP

MP

P

regions along the digit FIG. 6. Relative excursion between tendon and sheath for human fingers. Data are expressed per lo" of joint motion. Single joint motion: 0 MCP joint; E3 PIP joint; W DIP joint.

k 2.3 mm. Using the linear relationship, as shown in Eq. (I), the calculated excursion based on the data obtained from single joint rotations compared well with those measured in specimens with motion at all three joints (Fig. 7). The regression coefficient (3)was 0.95 and the slope was 1.05. The average variance was 5.0 1.0%. For the canine digit, the data for angular rotation, tendon excursion, sheath displacement, and tendon excursion relative to the sheath are summarized in Table 2. These specimens were tested under identical loading conditions as the human digits. In gen-

*

z 23 0

401

0 X W

ac

201

J

":/

y=105X-012

a

p < 0 0001

r2=0.95

0

1

0

0"

0

I

I

20

40

1

I

E X PER1 M EN T A L L Y MEASURED TOTAL EXCURS ION (mm)

FIG. 7. Comparison of total tendon excursion between experimentally measured values to calculated values [based on Eq. (l)]. Linear regression analysis showed good agreement between the two methods.

J Orthop Res, Vol. 8, No. 2, 1990

DISCUSSION A variety of methods have been used to measure tendon excursion subsequent to application of a known load (1,2,23). In most cases, however, only one of the phalangeal joints was allowed to rotate, and the tendon displacement was recorded in the region proximal to the joint. Also, the amount of joint rotation was arbitrary, i.e., measured without consideration of the COR. Since the phalangeal joints are not single-hinged joints, a goniometer measurement may not represent the true joint rotation. This error is compounded in the case of an intact digit, where multiple joint rotations take place with several CORs. In this study, we have determined the angular rotation and COR together with the displacements from the tendon excursion along the digit. According to a previous study (18), such a kinematic approach to measurement of joint rotations and tendon and sheath displacements results in a more representative description of joint kinematics. Therefore, we believe the results obtained in this study would be an important addition to previously published data. However, the method used to calculate the tendon excursion is only an approximation. The joint motion was assumed to be in one plane, but the actual motion of the digit may be a combination of axial rotation and lateral motion. Also, the tendon excursion in the area proximal to the joint in motion was defined as the straight-line distance between two positions. In the area distal to the joint, the tendon may bowstring. However, we assumed the tendon does not bowstring in this study, and calculated the distance between B' and B" shown in Fig. 3b. McGrouther and Ahmed (17) measured the FDP tendon relative to the sheath under passive motion in fresh cadaver and found 1.3 mm of tendon excursion for 10" of joint rotation. These results are sim-

173

FLEXOR TENDON EXCURSION

TABLE 2. The amounts of the tendon excursion, the sheath displacement, and the tendon excursion relative to the sheath of canine digits Single joint rotations

1. Angular rotation (deg) 2. Tendon excursion (mm) Regions C M PP P MP 3. Sheath displacement (mm) Regions C M PP P MP 4. Tendon excursion relative to the sheath (mrn) Regions C M PP P MP

M

PP

P

PIP joint

DIP joint

40.5 i 5.3

32.3 i 5.3

i 1.1 i 1.1 f 0.5 i 0.1 i 0.1

7.0 t 0.8 7.6 i 0.9 4.3 i 0.7 1.5 t 0.4 0.6 i 0.1

3.2 2.7 2.7 1.7 1.6

8.9 i 1.0 3.8 i 0.5 1.2 f 0.2 0.6 f 0.1 0.5 i 0.1

1.6 t 0.3 0.6 i 0.1 0.9 i 0.1 2.9 t 0.6 0.7 i 0.3

0.8 t 0.2 0.5 i 0.1 0.4 i 0.1 0.5 2 0.1 0.5 2 0.1

3.9 f 0.3 4.8 i 0.8 1.7 i 0.5 1.1 2 0.5 0.4 i 0.1

6.1 t 0.5 6.8 i 0.7 4.5 f 0.8 1.8 t 0.3 0.7 i 0.2

2.8 t 0.4 2.4 i 0.5 3.0 i 0.4 1.9 i 0.3 1.2 2 0.2

12.7 6.7 2.0 0.7 0.7

ilar to ours when 2 N was applied to the FDP tendon. This suggested that tendon motion under 2 N load was similar to that under passive motion. However, the load applied to the FDP tendon may have an effect on tendon excursion and the study on the relationship between the load and tendon excursion should be done in the future. The poor clinical results in the treatment of flexor tendon injuries are mainly due to scar formation between the flexor tendon and its surrounding sheath. Previous studies demonstrate that appropri-

C

MCP joint 71.5 i 5.8

MP

regions along the digit

FIG. 8. Relative excursion between tendon and sheath for canine digits. Data are expressed per lo" of joint motion. Single joint motion: 17 MCP joint; H PIP joint; H DIP joint.

2

0.5

i 0.5 i 0.6

t 0.4 i 0.3

ate mobilization post-operatively provides a means of minimizing such problems (6,7,8,10,11,14,15, 21,22). However, the amount of tendon excursion along the digir has been unknown. In this study, we have demonstrated that the tendon excursion in regions distal to the joint in motion is minimized (Fig. 5). Our data further suggest that flexion of joints distal to the repair site is more effective in achieving tendon excursion and therefore more effective in reducing scar formation after flexor tendon repair. Such findings should aid in the selection of locations for mobilization. The relative motion between the tendon and the sheath has been considered as an important factor to reduce scar formation (5,6,20,21). Therefore, large tendon excursions without their relative motion to the sheath may not be an effective treatment method. As in the case during MCP joint motion for region C , despite the large tendon excursion (2.1 mm/lO"), the tendon excursion relative to the sheath is small, i.e., only 0.8 mm/lO". On the other hand, during PIP joint motion, both the tendon excursion and tendon excursion relative to the sheath in zone I1 are large (2 mm/lO" for tendon excursion and 1.7 mm/lO" for tendon excursion relative to the sheath). Although the linear relationship between tendon excursion and angular rotation may be questioned, several previous studies supported the linear relationship (1,2,4). Therefore, a linear assumption is sufficient to compare the effectiveness of the indi-

J Orthop Res, Vol. 8 , N o . 2 , 1990

S. HORIBE ET AL.

174

vidual joint motions on tendon excursion relative to the sheath. Several animal species including canine, chicken, rabbit, and monkey have been used for tendon studies (7-11,15,16). To apply the results to clinical cases, it is important to compare the kinematics of these experimental animals to those of the human digits. The canine has a similar anatomy of the tendon apparatus as humans (7-10,18). Two flexor tendons enter the sheath proximally to the MCP joint in both species. Beyond the MCP joint, the superficial tendon is penetrated by the profundus tendon and inserts to the middle phalanx. The profundus tendon inserts to the distal phalanx. However, the pulley system is somewhat different. In humans, there are both cruciates and annular pulleys, whereas, in the canine, there are only annular pulleys (3). Also, the DIP joint of the canine is hyperextended. Despite these anatomical differences, the tendon excursion relative to the sheath within zone I1 is similar for the canine and human (Figs. 6 and 8). It is therefore concluded that the canine digit is a good experimental model for the study of flexor tendon kinematics within zone 11. Acknowledgment: Financial support from NIH grant AR 33097 and the Malcolm and Dorothy Coutts Institute for Joint Reconstruction and Research is greatly appreciated. We also thank Mrs. Laurie Aker for her assistance in editing this manuscript.

REFERENCES 1. An KN, Ueba Y, Chao EY, Cooney WP, Linscheid RL: 2.

3. 4.

5. 6.

Tendon excursion and moment arm of index finger muscles. J Biomeck 16:419-425, 1983 Armstrong TJ, Chaffin DB: An investigation of relationship between displacement of finger and wrist joints and the extrinsic flexor tendons. J Biomeck 11: 119-128, 1978 Brand PW, Cranor CKC, Rouge B, Ellis JC: Tendon and pulleys at the metacarpophalangeal joint of a finger. J Bone Joint Surg [Am] 57:779-784, 1975 Bunnell S: Repair of tendons in the fingers and description of two new instruments. Surg Gynecol Obstet 26:103-110, 1918 Cannon NM, Strickland JW: Therapy following flexor tendon surgery. Hand Clin 1:147-165, 1985 Duran RJ, Houser RG: Controlled passive motion following

J Orthop Res, Vol. 8, No. 2 , 1990

flexor tendon repair in zones two and three. In: American Academy of Ortkopaedic Surgeons, Symposium on Tendon Surgery in the Hand, St. Louis, C. V. Mosby, 1975 7. Gelberman RH, Amiel D, Gonsalves M, Woo SL-Y, Akeson WH: The influence of protected passive mobilization on the healing of flexor tendons: a biochemical microangiographic study. Hand 13:120-128, 1981 8. Gelberman RH, Manske PR, Akeson WH, Woo SL-Y, Lundborg G, Amiel D: Flexor tendon repair. J Orthop Res 4: 119-128, 1986 9. Gelberman RH, Manske RP, Vande Berg JS, Lesker PA, Akeson WH: Flexor tendon repair in vitro: a comparative histologic study of the rabbit, chicken, dog and monkey. J Orthop Res 2:3948, 1984 10. Gelberman RH, Woo SL-Y, Lothringer K, Akeson WH: Effects of early intermittent passive mobilization on healing canine flexor tendons. J Hand Surg 7:170-175, 1982 11. Hitchcock TF, Light TR, Bunch WH, Knight GW, Sartori MJ, Patwardhan AG, Hollyfield RL: The effect of immediate constrained digital motion on the strength of flexor tendon repairs in chickens. J Hand Surg 12590-595, 1987 12. Ketchum LD: Suture materials and suture techniques used in tendon repair. Hand Clin 1:43-53, 1985 13. Kleinert HE, Kutz JE, Ashbell JS, Martinez E: Primary repair of lacerated flexor tendon in “no-man’s land.” J Bone Joint Surg [Am] 49577, 1967 14. Lister GD, Kleinert HE, Kutz JE, Atasoy E: Primary flexor tendon repair followed by immediate controlled mobilization. J Hand Surg 2:441451, 1977 15. MacMillan M, Sheppard JE, Dell PC: An experimental flexor tendon repair in zone I1 that allows immediate postoperative mobilization. J Hand Surg 12582-589, 1987 16. Manske PR, Lesker PA: Histologic evidence of intrinsic flexor tendon repair in various experimental animals. An in vitro study. Clin Orthop 182:297-304, 1984 17. McGrouther DA, Ahmed MR: Flexor tendon excursions in “no-man’s land.” Hand 13:129-141, 1981 18. Spiegelman JJ, Woo SL-Y: A rigid-body method for finding centers of rotation and angular displacements of planar joint motion. J Biomeck 20:715-721, 1987 19. Spiegelman JJ, Woo SL-Y, Kwan MK: A mathematical expression for describing flexor tendon excursion across digital joints. 1986 Advances in Bioengineering, ASMEIBED 2~103-104, 1986 20. Strickland JW: Flexor tendon repair. Hand Clin 155-68, 1985 21. Strickland JW, Glogovac SV: Digital function following flexor tendon repair in zone 11. A comparison of immobilization and controlled passive motion techniques. J Hand Surg 5537-543, 1980 22. Woo SL-Y, Gelberman RH, Cobb NG, Amiel D, Lothringer K, Akeson WH: The importance of controlled passive mobilization in flexor tendon healing. A biomechanical study. Acta Ortkop Scand 52:615422, 1981 23. Youm Y, Gillespie TE, Flatt AE, Sprague BL: Kinematic investigation of normal MCP joint. J Biomeck 11:109-118, 1978