Outcome of zone 2 flexor tendon injuries: Kleinert versus controlled active motion therapy regimens

Abstract

Introduction

No single optimal regimen for the rehabilitation of flexor tendon injuries has yet been determined. We aimed to evaluate if a change in rehabilitation from a modified Kleinert to a controlled active motion (CAM) regimen had an effect on outcomes in a regional plastic surgery unit. We did this by comparing ruptures and range of movement of zone 2 repairs following both Kleinert and CAM regimens.

Methods

We performed a retrospective case series review, analysing data collected prospectively between 2004 and 2007. During 2004 and 2005, patients were rehabilitated with a modified Kleinert regimen, and during 2006 and 2007 a CAM regimen was used. We looked at total active motion (TAM) and ruptures at 12-week follow-up for all zone 2 repairs, and compared the two regimens.

Results

There were 38 patients with 42 injured digits in the Kleinert group, and 34 patients with 39 injured digits in the CAM group. There was no statistically significant difference in TAM achieved between the Kleinert and CAM regimens overall (70% versus 72% of normal in each group respectively, P = 0.70 t-test). Patients over 30 years old achieved significantly worse outcomes in the Kleinert group than in the CAM group (P = 0.03). One digit ruptured following a Kleinert regimen (2.6%) compared with four digits in the CAM group (11.7%).

Conclusion

In this study, we found no overall difference in outcome following a Kleinert or CAM rehabilitation regimen. Rupture rates were higher in the CAM group by four-fold. In our patients those over 30 years had poorer outcomes when rehabilitated with a Kleinert regimen.

Keywords

Flexor tendon Kleinert controlled active motion therapy regimen rehabilitation rupture

Introduction

In recent years a large body of research has focused on the postoperative rehabilitation of flexor tendons in an attempt to better understand the effects of postrepair motion stress on tendon strength and healing.^1–3 Several studies have demonstrated that early mobilization following flexor tendon injuries has advantageous effects including increased tendon excursion, decreased adhesion formation and increased tensile strength, including a recent systematic review.^1–5 Numerous regimens have been described; two of the commonest protocols fall into the categories of early passive mobilization and early active mobilization. In 1967, Kleinert et al.⁶ described a regimen of early passive mobilization involving active extension against rubber band resistance and passive flexion by the rubber bands.This regimen has been widely used demonstrating good results, and has been subject to several modifications over the years such as the addition of a palmar bar pulley.⁷

In an attempt to further improve outcomes, early active regimens – so-called ‘controlled active motion’ (CAM) methods were developed, with the patient actively flexing the involved digit from an early stage.⁸ While some studies have raised concerns regarding higher rupture rates with early active regimens, many others have demonstrated good results.^5,8–11 Various modifications of both these rehabilitation methods are in clinical use; however, no single optimum regimen for the rehabilitation of flexor tendons has been identified.⁵ Although a systematic review identified ‘weak evidence to support both early active motion protocols and protocols that combine passive mobilisation and active extension as providing superior results …’, it concluded that the definitive answer remains elusive and further research is required.⁵

In 2006, responding to advocates of a CAM regimen, the rehabilitation protocol following flexor tendon repairs changed in our regional hand unit from a modified Kleinert regimen to a CAM regimen. We aimed to establish if this change in rehabilitation regimen had an effect on our outcomes by comparing ruptures and range of movement of zone 2 repairs following both Kleinert and CAM regimens.

Methods

We collected data prospectively in the hand therapy department between 2004 and 2007, looking specifically at range of movement and ruptures at 12-week follow-up for all zone 2 flexor tendon repairs. We performed a retrospective analysis of these data to assess the consequences of the change in rehabilitation regimen during this time period, with all other variables including surgical technique, assessors and assessment methods staying constant. The study was performed as part of a service evaluation in our unit in accordance with the guidelines of our institution, and as such did not require Research Ethics Committee approval. Patients consented prospectively to the measurement and evaluation of their outcomes.

We included all patients presenting to the Regional Hand Unit between January 2004 and December 2007 with complete divisions of flexor digitorum profundus with or without flexor digitorum superficialis for all fingers, including single or multiple injured digits and concomitant digital nerve injuries.

We excluded thumb tendons, associated bony or joint injuries including fractures and revascularization injuries, and those patients who failed to attend 12 weeks of rehabilitation and follow-up.

Patients presenting between January 2004 and December 2005 were rehabilitated with a Kleinert regimen as per our department's protocol. Those presenting from January 2006 to December 2007 were rehabilitated with a CAM regimen as per our department's protocol.

Kleinert regimen

We used a Kleinert regimen of active extension with passive flexion using rubber bands, incorporating modifications developed by May et al.⁷ The splint was fitted days 3–5 postoperatively with the wrist in neutral and the metacarpophalangeal joints (MCPs) in 50–70° of flexion. Traction was applied to all fingers and passed through a palmar pulley. Traction was released overnight with the fingers resting in extension. Hourly exercises included an active hold for 2–3 seconds. Traction was removed at week 3 and gentle active flexion commenced. The splint was removed completely at week 5 with full function permitted by week 12.

CAM regimen

We used a modified CAM regimen of active flexion and extension within a dorsal splint similar to that described by Small et al. in 1989.⁸ The splint was fitted days 3–5 postoperatively with the wrist in neutral and the MCPs in 50–70° of flexion. Exercises were performed every two hours to include passive flexion of all fingers, full active extension of interphalangeal joints to the dorsal splint and gentle active flexion. The splint was removed at week 5 with full use of the hand permitted at week 12.

During both regimens patients attended a weekly appointment in the Hand Unit between weeks 1–6, with further weekly appointments until 12 weeks postinjury depending on progress. Patients were given a further appointment at 26-week follow–up; however, failure to attend was extremely high by this stage, and so these results were not included in this study.

Surgical technique

We used a two-strand modified Kessler core suture of 3-0 or 4-0 Prolene with a circumferential 6-0 suture for all repairs.

Outcome measures

The outcome measures we looked at were rupture rate and total active motion (TAM) at 12 weeks (MCP + PIP + DIP flexion minus extensor lag) expressed as a percentage of the same measurement taken from the contralateral, uninjured (normal) digit.⁹ These were categorized as excellent (TAM = 100%), good (TAM 75–99%), fair (TAM 50–74%) or poor (TAM < 50%). Measurements were performed by one of a team of four trained senior hand therapists using a standardized protocol and the same goniometer to assess within 2° accuracy.

Statistical analysis

Analysis of the two groups was performed using Microsoft Excel, using parametric (t-tests) and non-parametric tests (Mann-Whitney U tests, χ ² test) as appropriate. Correlation analysis was performed using Pearson's coefficient. To avoid bias from the patients with multiple injured digits, such patients had the mean outcome of each injured digit calculated to allow analysis to be based on patient rather than digit.

Results

There were 38 patients and 42 digits in the Kleinert group, and 34 patients and 39 digits in the CAM group. Both groups had a male predominance with a median age of 29 and 34 years, respectively. There were more dominant hands injured in the CAM group, although this did not reach statistical significance (59% versus 35%, P = 0.06, χ ²). There was no significant difference between the two groups in terms of which digit was involved (P = 0.51), nor in terms of which tendon was involved (P = 0.55). There was no significant difference between the two groups in terms of concomitant nerve injury (P = 0.12). There was no difference in waiting time for surgery, with most patients in both groups experiencing between one and three days delay. Table 1 illustrates the demographics of the two groups.

Table 1

Patient demographics in the two treatment regimens

	Kleinert	CAM	P value
Patients (n)	38	34
Digits injured (n)	42	39
Age (median, range)	29 (13–63)	35 (18–82)	0.23 (MWU)
Males (n)	23	23	0.53 (χ ²)
Dominant side injured	16 (35%)	23 (59%)	0.06 (χ ²)
Tendon injured
FDS only	1 (2%)	0 (0%)	0.55 (χ ²)
FDP only	17 (40%)	19 (49%)
FDS and FDP	22 (52%)	20 (51%)
Not recorded	2 (4%)	0 (%)
Digit
Little	17 (40%)	13 (33%)	0.51 (χ ²)
Middle	12 (29%)	8 (21%)
Ring	7 (17%)	8 (21%)
Index	6 (14%)	10 (26%)
Associated nerve injury	20/42 (48%)	12/39 (31%)	0.12 (χ ²)
Days to surgery (mean)	1.57	2.6

CAM, controlled active motion; FDS, flexor digitorum superficialis; FDP, flexor digitorum profundus; MWU, Mann-Whitney U test; χ², chi-square test

As well as both groups being similar at baseline in terms of demographic characteristics, outcomes in both groups, measured by TAM, were not significantly affected by how many digits or which digit the patient injured, whether it was their dominant hand injured, or the presence of a concomitant digital nerve injury.

There was no statistically significant difference in TAM achieved between the Kleinert (70% normal, range 35–100%) and CAM groups as a whole (72% normal, range 21–98%, P = 0.70, t-test). Post hoc power analysis demonstrated that 672 patients would need to be recruited to demonstrate a significant difference; however, 42% of the Kleinert group were excellent or good, and 47% of the CAM group were excellent or good at 12-week follow-up (Table 2).

Table 2

Outcomes measured by total active motion (TAM) and ruptures in the two rehabilitation groups at 12-week follow-up

	Kleinert	CAM	P value
Number	38	34
Total active motion (°)
Mean	177°	182°	0.73 (t-test)
Range	92–240°	56–254°
Percentage of normal
Mean	70%	72%	0.70 (t-test)
Range	35–100%	21–98%
Age
Good/excellent in <30 years	13/38 (34%)	7/34 (21%)	0.20 (χ ²)
Good/excellent in >30 years	3/38 (8%)	9/34 (26%)	0.03 (χ ²)
Age
Correlation with % TAM (r)	−0.43	−0.31
Rupture	1/38 (2.6%)	4/34 (11.7%)	0.18 (χ ²)

Age appears to have had an adverse effect on outcome in the Kleinert group, but not the CAM group. When considering only those patients less than 30 years old there was no significant difference between the outcomes of the Kleinert or CAM groups. However, when considering those patients over 30 years old the Kleinert patients did significantly worse than the CAM group, with 92% of those rehabilitated with a Kleinert regimen demonstrating poor or fair range of movement (P = 0.03). This is further demonstrated by a negative correlation between percentage recovery of TAM and increasing age in the Kleinert, but not the CAM group (r = −0.43, Figure 1).

Figure 1

Negative correlation between increasing age and % recovery of total active motion (TAM) in the Kleinert group

There was one rupture in the Kleinert group giving a rupture rate of 2.6% of patients. There were four ruptures in the CAM group giving a rupture rate of 11.7% of patients (P = 0.18, χ ² test). These ruptures were single digit ruptures in different patients. The reasons documented for rupture included failure to comply with splintage, accidental fall or spontaneous rupture. These outcomes are shown in Table 2. The mean age of the patients with ruptures in the CAM group was slightly older than the Kleinert rupture (33 versus 25 years), although the numbers are too small for statistical analysis.

Discussion

In this study, we found no overall difference in range of movement achieved following a Kleinert or CAM rehabilitation regimen at 12 weeks post zone 2 flexor tendon repair. A higher rupture rate was seen in the CAM group (11% versus 2%), corresponding to a clinical increase in ruptures by four-fold. Concerns about a possible higher rupture rate with CAM regimens have previously been raised in the literature with rates as high as 46% reported.¹² The findings of our study are comparable to those in Peck et al.'s¹² study that looked at a similar demographic group of patients to ours and found no difference in outcomes between Kleinert and CAM but a significantly higher rupture rate in the CAM group. Poor compliance was an issue in their patients, and there is no doubt we face a similar problem in our unit. Bainbridge et al.¹³ also reported comparable rupture rates to our study but better outcomes in their zone 2 flexors rehabilitated with a CAM regimen.

Beyond the literature concerning paediatric injuries, there is little in the literature identifying differences in outcomes of flexor tendon repairs by patients’ age. Within our patient cohort, approximately half the patients we see fall into the under 30-year-old age group. Anecdotally we have noticed particularly poor compliance in this patient group, and for this reason decided to analyse their outcomes independently in this study. We saw in fact that those over 30 years had poorer outcomes when rehabilitated with a Kleinert regimen. We might hypothesize that these patients have a higher risk of co-morbidities such as osteoarthritis leading to an increased tendency to stiffness following a more restrictive regimen. Alternatively it might be increased compliance with treatment in the over 30s, who conform more rigidly to the restrictions of such a regimen than their younger counterparts, that leads to increased stiffness. The mean age of the patients with ruptures in the CAM group was slightly older than the Kleinert rupture. Although the numbers are too small for statistical analysis, it appears that the pay-off for better range of motion (ROM) achieved with the CAM regimen in older patients may be an increased rupture rate. Further investigation into a correlation between age and outcomes is warranted.

The strengths of this study are its prospective data albeit retrospectively analysed and standardized methods, with surgical technique, assessors and assessment methods staying constant. In keeping with this standardization we used a two-strand repair for all tendons in this study group having found this repair easy to teach and learn as a unit standard. Evidence suggesting a potentially negative biological effect of multistrand locking repairs and the benefits of decreased tendon handling, as well as a previously audited and acceptably low rupture rate, supported our view to continue with the two-strand repair while evaluating the effect of changing the rehabilitation regimen.¹⁴

Several studies, however, have shown that the strength of a tendon repair is proportional to the number of suture strands that cross the repair site.^15–17 This has initiated a move from two-strand core suture techniques and the emergence of a number of four-strand or more techniques in clinical practice. As illustrated by Strickland,⁴ any four-strand core suture repair combined with a running lock or horizontal mattress circumferential epitendinous suture should permit light composite digital flexion during the entire healing period, and is therefore appropriate for early active mobilization regimens. A recent systematic review of flexor tendon rehabilitation protocols in zone 2, however, showed no relationship between the number of core strands and rupture rate.⁵

Although our study did not demonstrate statistically significant differences in range of movement outcomes between the two rehabilitation regimens overall, we do feel they could be further improved. To this end we are now looking at adopting a four-strand repair alongside the CAM regimen, and propose future study to evaluate the efficacy of this combination.

We acknowledge our limitations, in particular, the small study group size, which may have prevented statistically significance results being uncovered. While many attempts were made to standardize the study, several different surgeons performed the surgical repairs, albeit to a unit standard technique as described. The majority of repairs were performed by registrar-grade surgeons in both groups. Patient compliance could also not be objectively assessed, and neither could attendance at follow-up be controlled. We also acknowledge that the Strickland system of evaluating the outcome of range of motion is considered to be among the most rigorous classification system and may have provided a better basis for comparison with other studies.⁵

In conclusion, zone 2 flexor tendon injuries remain challenging injuries to repair and rehabilitate effectively. While debate continues regarding the most appropriate method of rehabilitating flexor tendon injuries it seems most likely that a ‘one technique suits all’ policy will remain elusive and we must be prepared to moderate our protocols based on individual patients’ requirements. A Kleinert regimen remains suitable for many patients, demonstrating a low rupture rate. It may be particularly suitable for patients thought to be at increased risk of rupture. Patients over 30 years old, however, may benefit most from a CAM regimen alongside an appropriately strong tendon repair.

Competing interest: None declared.

References

Thien

, Becker

, Theis

J-C

. Rehabilitation after surgery for flexor tendon injuries in the hand. Cochrane Database Syst Rev 2004, Issue 4. Art No: CD003979. DOI: 10.1002/14651858.CD003979.pub2

Gelberman

, Amiel

, Gonsalves

, The influence of protected passive mobilisation on the healing of flexor tendons: a biological and microangiographic study. Hand 1981;13:120–8

Pentengill

. The evolution of early mobilization of the repaired flexor tendon. J Hand Ther 2005;18:157–68

Strickland

. The scientific basis for advances in flexor tendon surgery. J Hand Ther 2005;18:94–109

Chesney

, Chauhan

, Kattan

, Systematic review of flexor tendon rehabilitation protocls in zone 11 of the hand. Plast Reconstr Surg 2011;127:1583–92

Kleinert

, Kutz

, Ashbell

, Primary repair of lacerated flexor tendons in no-man's land. J Bone Joint Surg Am 1967;49:577

May

, Silfverskiold

, Sollerman

. Controlled mobilization after flexor tendon repair in zone II: a prospective comparison of three methods. J Hand Surg 1992;17A:942–52

Small

, Brennan

, Colville

. Early active mobilization following flexor tendon repair in zone 2. J Hand Surg 1989;14B:383–391

American Society of Surgery of the Hand. Clinical Assessment Committee Report. Rosement, IL: Churchill Livingstone, 1976

10.

Cullen

, Tolhurst

, Lang

, Page

. Flexor tendon repair in zone 2 followed by controlled active mobilization. J Hand Surg 1989;14(B);396–9

11.

Cetin

, Dincer

, Kecik

, Rehabilitation of flexor tendon injuries by use of a combined regimen of modified Kleinert and modified Duran techniques. Am J Phys Med Rehabil 2001;80:721–8

12.

Peck

, Bucher

, Watson

, Roe

. A comparative study of two methods of controlled mobilization of flexor tendon repairs in zone 2. J Hand Surg 1998;23B:41–5

13.

Bainbridge

, Robertson

, Gilles

, Elliot

. A comparison of postoperative mobilization of flexor tendon repairs with ‘passive flexion-active extension’ and ‘controlled active motion techniques’. J Hand Surg 1994;19B:517–21

14.

Wong

, Cerovac

, Ferguson

, Mcgrouther

. The cellular effect of a single interrupted suture on a tendon. J Hand Surg (Brand Eur Vol) 2006;31:358–67. DOI: 10.1016/J.JHSB.2006.03.162

15.

Komanduri

, Phillips

, Mass

. Tensile strength of flexor tendon repairs in a dynamic cadaver model. J Hand Surg (Am) 1996;21:605–11

16.

Savage

, Risitano

. Flexor tendon repair using a ‘six strand’ method of repair and early active motion. J Hand Surg (Br) 1989;14:396–9

17.

Strickland

. Flexor tendon injuries: II. Operative technique. J Am Acad Orthop Surg 1995;3:55–62