Relative efficacy of three different tendon repairs in complete flexor digitorum profundus laceration in Zone I: A randomized controlled study

Hand flexor tendon injuries are common and biomechanically challenging to achieve good functional outcomes. Several approaches using the Pennington‐modified Kessler repair technique have been attempted, but high‐level evidence is still lacking. Here, we evaluated the relative efficacy of three versions of the Pennington‐modified Kessler technique in repairing complete flexor digitorum profundus (FDP) laceration in Zone 1. We conducted a 2‐year, single‐center, double‐blind, randomized clinical trial involving 85 patients with 105 digits enrolled between June 1, 2017 and January 1, 2019. Eligible participants were 20–60 years of age and underwent tendon repair in the acute phase for complete FDP laceration distal to the insertion of the superficial flexor tendon. The digits were randomized 1:1:1 to three treatment groups: (1) Pennington‐modified Kessler repair; (2) Pennington‐modified Kessler repair followed by circumferential tendon suture; or (3) Pennington‐modified Kessler repair followed by circumferential epitenon suture. The primary endpoint was total active range of motion (TAROM) at 2 years after the initial surgery. The secondary endpoint was the reoperation rate. Compared with group 1, both techniques for peripheral suture were associated with a decrease in TAROM at 2 years after surgery. The total reoperation rates of the three groups were 11.4%, 18.2%, and 17.6%, and we found no significant differences among the three groups possibly due to the limited sample size. Unexpectedly, among participants with complete FDP laceration in Zone I, both circumferential‐tendon and circumferential‐epitenon sutures caused worsening of TAROM after 2 years. No conclusions can be drawn regarding reoperation rates among the groups. Level of evidence: Therapeutic level I.

Injuries involving the flexor tendons in the hand are common and mainly affect young and active persons, whose cases comprise up to 10% of all hand injuries. 1,2Hand flexor tendon injuries have a tremendous impact on working ability and activities of daily living by impairing hand strength and dexterity. 3,4Despite the common nature of flexor tendon injuries, they present a major challenge to hand surgeons, not in the surgical treatment per se but in achieving a good functional outcome after surgery. 5rly active motion is the cornerstone for functional recovery after flexor tendon injuries. 6Many clinical studies have shown that early active motion after flexor tendon repair prevents adhesion formation and improves the functional outcome. 7,8However, early active motion elicits additional biomechanical stress on tendon repair to resist gap formation and rupture. 9Core suture techniques, including Pennington-modified Kessler sutures, have been developed to meet these biomechanical requirements. 10Although, to date, there is no high-level evidence identifying the optimal number of core strands and suture configuration, a meta-analysis revealed that the Pennington-modified Kessler technique is the most widely used technique in the currently available literature. 11re and peripheral sutures are typically used together in clinical practice. 12Biomechanical studies have revealed that the peripheral part is the weaker component after core sutures and initiates repair failure. 13,14sed on these biomechanical data, peripheral sutures have been frequently recommended because they contribute significantly to the biomechanical strength of tendon repair. 15However, the efficacy of peripheral sutures remains inconclusive because no high-quality clinical evidence has been reported. 5In addition, peripheral sutures could involve a 2-3 mm depth from the tendon surface or epitenon only. 14,16The aim of this study was to evaluate the relative efficacy of three versions of the Pennington-modified Kessler technique in repairing complete flexor digitorum profundus (FDP) lacerations in Zone I.

| MATERIALS AND METHODS
This study was a 2-year, double-blind, three-arm, single-center randomized clinical trial.The study protocol was approved by institutional ethics committees.Informed consent was obtained from all donors in accordance with the Declaration of Helsinki.The work has been reported in line with Consolidated Standards of Reporting Trials (CONSORT) Guidelines. 17

| Participants
Eligible participants were aged 20-60 years old; had complete FDP laceration distal to the insertion of the superficial flexor tendon (Zone I injuries); and were receiving tendon repair in the acute phase (less than 24 h after injury).Exclusion criteria included history of tendon injury; history of brain injury/disease; accompanying upper limb fractures, vascular, and nerve injury of the affected finger(s); additional upper limb tendon injury (flexor digitorum superficialis [FDS]); history of malignant tumors; presence of skin defects requiring flap coverage; mental illness history; and physical disability (Figure 1).All demographic parameters, including age, sex, insurance type, education level, and smoking status, were self-reported by patients.Injury mechanisms were also self-reported and classified into two categories (blade and saw) by researchers.

| Randomization and interventions
Patients were enrolled from June 1, 2017 to January 1, 2019.The last participant visit during the 2-year follow-up analyses occurred on December 29, 2020.Randomization was based on digits instead of participants.Each digit in this study received randomization and was considered one "case."Computer-generated randomization was performed in block sizes of three, stratified by injury pattern (single-digit, double-digit, or three-or-more digit injury).Digits with FDP laceration were randomized 1:1:1 to three treatment groups: (1) Pennington-modified Kessler (four-stranded 3-0 core suture) repair, (2) Pennington-modified Kessler (four-stranded 3-0 core suture) repair followed by circumferential tendon suture (5-0 Prolene), and (3) Pennington-modified Kessler (four-stranded 3-0 core suture) repair followed by circumferential epitenon suture (6-0 Prolene).All the surgery were completed under branchial plexus nerve block anesthesia by a group of senior specialists in hand surgery.The bite sizes of Pennington-modified Kessler repair, circumferential tendon suture and circumferential epitenon suture were 7.5, 2, and 1 mm, respectively.The circumferential tendon suture had a purchase 2 mm deep from the surface of the tendon. 16e three repair techniques are illustrated in Figure 2. Simple-running sutures were used for peripheral sutures.The A4 pulley was released.
The Kleinert protocol was used as the general principle in this study.
Briefly, the rehabilitation of fingers began as soon as possible after surgery.A Kleinert dynamic flexion splint with a rubber band was used for 4 weeks (Figure 3).Since the purpose of the splint is to serve as a cautionary reminder rather than maintaining the wrist and fingers in specific positions, a short forearm splint suffices. 18Considering the variation in injury and patient factors, a multidisciplinary team (MDT) consisting of physiotherapists and surgeons adjusted the rehabilitation protocol.The protocol after tendon repair followed the general principles described in previous reports and was personally adjusted for each digit. 18,19The progressive exercises program commonly included passive and active joint motion, light functional activities, and musclestrengthening exercises.The level of activity difficulty was increased based on patient tolerance, adhesions, and previous progress, other than on a strict timeline.When MDT is considered appropriate based on the progression of rehabilitation, out-of-splint motion with the wrist in a functional position would be allowed in these patients. 18Before discharge, each participant will learn the outlines of the basic Kleinert protocol after detailed explanation.Secondary surgeries (e.g., tenolysis and capsulotomy), if needed, were typically performed 6 months after the initial surgery.

| Blinding
In this study, only the treating surgical team (surgeon, nurse, and anaesthesiologist) knew the group information.The nurse obtained treatment information from an interactive system and then showed this information to the treating surgeons.The treating surgeons then completed the operation using the requested techniques.Patients and other investigators were blinded to the group information.

| Primary endpoint
The primary endpoint was the total active range of motion (TAROM).The ranges of motion for active metacarpophalangeal and proximal and distal interphalangeal joints were measured and summed to calculate the TAROM at 2 years after surgery.The examiners measuring TAROM were blinded to treatment group information.

| Secondary endpoints
The secondary endpoint was the reoperation rate.Fair (50%-74% of the contralateral side) or poor (<50% of the contralateral side) TAROM according to the standard proposed by the International Federation of Societies for Surgery of the Hand after tendon repair was indicated for tenolysis with or without capsulotomy. 20Secondary rupture after initial repair is indicated for reoperation and is commonly treated with tendon reconstruction.
The minimal clinically important difference for TAROM and the primary endpoint is unknown.The sample size was based on type I error (α) set at 0.05, with 80% power (1 − β).Based on the data obtained from the pilot study, 31 digits were required for each group to detect 20°differences, and we finally set the sample size to 35 digits after considering a possible 10% loss to follow-up.

| Statistical analysis
Statistical analysis was performed based on the assumption that Pennington-modified Kessler (3-0) repair was the control, and the Dunnett method was used for multiple comparison adjustment.To compare TAROM, the primary endpoint in this study, among the three groups, we performed linear regression adjusted for age, injury pattern, and injury mechanism.Adjusted mean differences and 95% confidence intervals (CIs) were calculated with a linear regression model.Secondary endpoints and adverse events were analyzed with Fisher's exact test.Categorical and continuous variables are expressed as the mean ± standard deviation and count (percentage).To be conservative, as prespecified in the statistical analysis plan, the analysis was based on an intention-to-treat (ITT) population.In this study, the ITT population was consistent with the population grouped by actual treatment.Because of the potential for type I error due to multiple comparisons, findings for the analyses of secondary endpoints should be interpreted as exploratory.Bonferroni correction was used.

| RESULTS
A total of 85 patients with 105 digits were randomized between June 1, 2017 and January 1, 2019.No patient was unable or unwilling to provide informed consent after randomization.In this study, the ITT population was consistent with the population grouped by actual treatment.The baseline characteristics of the patients were similar across study groups (Table 1).
Unexpectedly, compared with the control treatment with Pennington-modified Kessler repair (group 1), both Penningtonmodified Kessler repair followed by circumferential tendon suture (group 2) and Pennington-modified Kessler repair followed by circumferential epitenon suture (group 3) were associated with a decrease in TAROM at 2 years after surgery.When the digits received add-on circumferential tendon sutures (group 2), the mean adjusted difference in TAROM was −24.6°(95%CI: −10.0°to −39.2°).

| DISCUSSION
In this study, we evaluated the relative efficacy of three versions of the Pennington-modified Kessler technique in repairing complete FDP ruptures.The results showed that, among participants with complete FDP laceration in Zone I, both circumferential-tendon and circumferential-epitenon sutures caused worsening of TAROM.For instance, Lotz et al. noticed that more than 60% of the load was carried by the peripheral suture at the point of its rupture, 21 indicating the value of peripheral sutures.Thus, traditionally, tendon repair has been defined as a composite of the core and peripheral sutures.We assumed that most hand surgeons would repair the flexor tendon by using core and peripheral sutures together.Our study demonstrates that the strength of modified Kessler tenorrhaphy is already sufficient for early mobilization and active rehabilitation programs after FDP repair.More strikingly, the two groups with add-on peripheral sutures had worse outcomes in terms of TAROM.But it is worth mentioning that, considering that if the reoperated patients/digits are excluded, the adverse effects of different suture methods on reoperation rate and TAROM will be weakened, the analyses of TAROM were performed including them.

Group
What's more, the results remain the same in this study if the reoperated digits are excluded.
In this study, we included patients with complete FDP laceration distal to the insertion of the superficial flexor tendon.This is easy for study design because the tendon injuries in Zone II are more complicated and could involve FDPs, FDSs, or FDPs/FDSs together.
Because the main goal of this study was to compare the efficacy of different peripheral sutures, we obtained sufficient statistical power based on a smaller sample size by only including Zone I cases.
The role of postoperative immobilization after flexor tendon repair is controversial.Clearly, prolonged postoperative immobilization after FDP repair might lead to adhesions and compromised outcomes.However, early active and passive motion could also lead to the development of a gap at the tenorrhaphy site and an increased possibility of secondary rupture.The recommended period of postoperative immobilization ranges from 0 to 3 weeks. 5In this study, we used our traditional protocol in which a Kleinert dynamic flexion splint with a rubber band was used for 4 weeks during postoperative rehabilitation.A pyramid of progressive exercises program 22 was used based on individual tissue responses to determine when to advance therapy.As we described in the Methods, this program is based not on a strict timeline but on adhesions and the tensile strength of the repair.
Improved outcomes of primary flexor tendon repair have greatly diminished the need for reoperation.The overall reoperation rate after FDP repair in this study accounted for less than 8% of all operations after excluding those reoperations due to infection in this study.This reoperation rate was similar to those reported by previous studies. 23Infections are rare after tendon repair when debridement and irrigation were sufficient during the surgery procedure. 24In this study, only nine superficial infections occurred evenly in the three groups and there were no significant differences among them.This extent of infection did not largely hinder the recovery process after the therapist adjusted their rehabilitation protocol.
To maximize the consistency between samples, all the participants received primary repair within 24 h.Although prolonged delay may permit unacceptable retraction of tendons, such retraction is not common in Zone I. So, we think the study findings can also be extrapolated to so-called delayed primary repair (less than 2 weeks).
This study has several limitations.First, because the incidence of secondary endpoints was very low, the secondary endpoints were not adjusted for confounding factors.Second, due to the 24-month observation duration, the magnitude of observed structural changes may not have represented clinically meaningful benefits to participants during the follow-up periods of this study.Finally, since circumferential tendon suture and circumferential epitenon suture were conducted on different tissues, we used 5-0/6-0 Prolene according to the toughness, respectively.This might confound the results about bite depth to certain extent.

| CONCLUSION
In this study, both the circumferential tendon suture and circumferential epitenon suture yielded worse TAROM than Penningtonmodified Kessler repair in digits with complete FDP laceration in Zone I.However, no conclusions can be drawn regarding reoperation rates among the groups because the study was not powered to evaluate this endpoint.

F
I G U R E 3 A dorsal blocking splint was used in Kleinert protocol.Passive flexion was achieved by an elastic band applied to fingernails (left).Active extension was permitted within the limits of the blocking splint (right).F I G U R E 2 The illustration of different repair techniques.ZHU ET AL.|2325 Comparison of primary and secondary endpoints versus group 1 for each digit.
Abbreviation: SD, standard deviation.TA B L E 2Abbreviations: AMD, adjusted mean difference; CI, confidence interval; SD, standard deviation; TAROM, total active range of motion.