Biomechanical Strength Comparison Between Pulvertaft Weave and Side-to-Side Tendon Repair: A Meta-Analysis and Systematic Review

Introduction

Tendon repair in the hand has been a continuously developing topic in the field of orthopaedic surgery. Various properties and types of tendon repair have been studied throughout the years and in conjunction with this, postoperative protocols or rehabilitation methods have been influenced as well. Hand surgeons should be knowledgeable about the possible choices for tendon repair and postoperative care that follows. For this study, the authors focused on the concept of early mobilization after tendon repair. Studies have shown that early mobilization can contribute to the reduction of adhesion formation and reduction of postoperative recovery time is possible by improving the range of motion and tensile strength of the tendon.[1–4] On the other hand, excessive strain placed on the repaired tendon during motion can also lead to rupture. One of the factors that could help with this dilemma is the strength of the tendon repair construct.[5,6]

The Pulvertaft (PT) weave has been one of the more widely used techniques for tendon repair and considered by some as the “gold standard.” Classically, the technique involves the distal tendon woven through three incisions in the proximal tendon and these weaves are inserted orthogonally before sutures are placed.[7–9] The side-to-side (STS) tendon repair provides us with another technique that is said to show properties that are equal or even greater than the PT while being less cumbersome in technique.[5,10,11]

Methodology

Search strategy and selection criteria

We conducted a comprehensive literature search using PubMed, Cochrane Library and Google Scholar from August 2020 to November 2020. Electronic searchers were performed and included: “Tendon” AND “Repair” AND “Graft*” AND/OR “Pulvertaft”

After the search was done, the authors then selected possible articles to be reviewed. The titles and abstracts were read to classify possible articles. Articles considered were included if they met the inclusion criteria; clinical trials/experimental studies, in vitro or in vivo, comparison of the PT weave to the STS and outcomes that compared biomechanical properties. Potential articles were then retrieved and screened by way of full text eligibility. In addition, the reference section of each potential article was reviewed to identify other eligible studies not captured by the initial search.

Assessment of Risk of Bias

Two investigators assessed the risk of bias of each included study independently. In the instance of inconsistency between the two, the research adviser would resolve the conflict and come up with a consensus. Risk of bias was assessed using the Risk of Bias In Non-Randomized Studies of Interventions (ROBINS-I). As in the tool for randomized trials, risk of bias was assessed within specified bias domains, and review authors were asked to document the information on which judgments were based. The studies were assessed as low risk, moderate risk, serious risk, critical risk or no information according to the following domains: confounding variables, selection of participants into the study, classification of interventions, deviations from intended intervention, missing data, measurement of outcomes and selection of reported results.

Data Extraction and Analysis

Statistical analyses were conducted using STATA MP Statistical Software, Version 13, College Station, TX: StataCorp LP. A p-value ≤0.05 was considered statistically significant. Since this study did not assume one effect size among all the studies, the overall effect of the meta-analysis was derived using a random-effects model (REM), which takes within-study and between-study variation into account. However, should a study have no heterogeneity, analysis was conducted using fixed-effects model (FEM). Standardized mean difference was used to pool the effect measure of continuous-level outcomes (maximum load and load to failure).

Statistical heterogeneity between studies were scrutinized using Q statistics test, I² statistics and tau squared (τ²) statistics (Higgins, 2003). I² values greater than 50% imply substantial heterogeneity (Higgins, 2003). Publication bias was assessed using both graphical and statistical approaches. Graphical approach for publication bias included contour-enhanced funnel plots, while statistical assessment of funnel plot asymmetry was performed using Begg’s asymmetry test (Higgins, 2003).

Results

The literature search yielded a total of 624 results. After duplicates and articles were screened, there were 53 articles for full text eligibility. The remaining six studies were included in this meta-analysis and systematic review (Table 1). All the six studies compared PT and STS. One study also compared the two techniques to another technique, loop technique but was still included in this analysis. All studies were done in vitro. With respect to the source of tendons, one study used turkey tendons, another used sheep, two studies used porcine, and two studies used human cadavers. Among the studies, all provided details on how they proceeded to do the repair technique. The risk of bias for the studies ranged from low to moderate risk. This can be seen in more detail in Table 3. As of the search of this study, there were no local studies found.

Pooled Estimate for Maximum Load Between Side-to-Side and Pulvertaft Weave Tenorrhaphy

Figure 2 illustrates the pooled summary effect for maximum load between STS and PT weave tenorrhaphy. A total of three eligible and complete studies were included, with a total of 36 specimens in the STS approach and 37 specimens in the PT weave tenorrhaphy. Results showed that there was no evidence that the two tendon repair techniques were significantly different (SMD = –0.84, z = 0.88, p = 0.379, 95% CI = –2.72 – 1.04). However, there was significantly high heterogeneity detected among the included studies (Q=21.10, p = 0.001, I² = 90.50%, τ² = 2.47).

Pooled Estimate for Load to Failure Between Side-to-Side and Pulvertaft Weave Tenorrhaphy

The pooled estimate for the load to failure outcome between the two tendon repair techniques is depicted in Figure 3. It can be noted that a total of four complete studies were included, involving a total of 83 specimens in the STS approach and 85 specimens in the PT weave tenorrhaphy. Analysis indicated that the load to failure was statistically higher in the STS approach (SMD = 1.36, z = 5.26, p = 0.001, 95% CI = 0.85 – 1.86) than the PT approach. It is also notable that a small, non-significant heterogeneity was detected among the included studies (Q=3.66, p = 0.300, I² = 18.10%, τ² = 0.05).

Publication Bias Assessment

The graphical analysis of publication bias using contour-enhanced funnel plots are presented in Figure 4. It can be noted in the funnel plot for maximum load that there is no evidence of funnel asymmetry; however, the funnel plot for load before failure does indicate the likelihood of funnel asymmetry. Nonetheless, statistical analyses using Begg’s Test and Egger’s Test (Table 2) suggest that publication bias among the included studies were unlikely.

Discussion

The PT weave is the standard for tendon grafting due to its strength and reliability. As such, it has been subjected to numerous studies that test its different properties with various modifications.[12–14]Recent studies have also compared it to other techniques in order to make up for its disadvantages.[5,6,11,13,15–17] These would include its bulkiness, the amount of tendon length needed to create the construct and the technique itself is demanding. One of these new techniques that is being compared to the gold standard is the STS tendon repair. It is a technique that was created to make up for some of the limitations of the PT weave while not comprising tensile strength. To the knowledge of the authors, this is the first meta-analysis and systematic review comparing the biomechanical properties of the STS technique to the PT weave, more specifically the maximum load and load to failure of the tendon repair techniques. As defined, maximum load is the peak force achieved in tensile testing while load to failure is the first negative inflection of force during the failure test.

All studies explained how the techniques were done. The standard PT weave technique was followed by most studies, however, the study by Kannan modified the technique because of standardization. For the STS repair, two studies deviated from the standard STS technique. Kannan, et.al. added sutures to both donor and recipient tendon. Bidic, et.al., on the other hand, did not incorporate a weave into the STS technique and sutured both overlapping tendons together.

Maximum load was found not to be significant in this study. This can be attributed to the variation in technique that the studies used. As mentioned, one study sutured the tendons in overlapping sites, but the donor tendon did not pass through a weave of the recipient tendon.

For the maximum load, there was considerable heterogeneity found between the studies while for the load to failure, only small amount of heterogeneity was found. To further analyze heterogeneity, a subgroup and meta-regression analysis was attempted, however, due to the small sample size, it was not possible. A funnel plot was then created to detect if there was any publication bias that could be the cause of heterogeneity. The funnel plots showed for both outcome measures that publication bias was not the case. Additionally, a Begg’s test and Egger’s test confirmed this. Aside from the limitations of this study, possible causes of heterogeneity are the small number of studies available and thus, the small sample size as well.

One of the evidences that display the STS having a higher load to failure could be the site of failure. Most of the studies noted their observation on how the tendon graft failed. For the PT, these included slipped suture knots or the donor tendon pulled through the recipient tendon. For the STS, these included longitudinal shear of the tendon from the repair site and failure of the site as well. One could attribute it to the number of weaves of the PT because the STS only has one weave. However, a recent study by Choke (2019), investigated the difference in strength of a PT weave with varying number of weaves. They concluded that the two weaves are a suitable number.

One limitation to this study is their in-vitro nature of the studies. This would let the researchers test the outcomes measurable by a tensile testing machine of the repair technique. It does not consider other important factors of tendon healing such as friction, adhesion formation and vascular healing properties provided by the tendons which can be provided by studies in vivo. Another weakness is that while most of the studies followed the protocol for the technique, others made modifications. Lastly, the source of tendons can also be a factor. However, there are studies that have shown that porcine, sheep and turkey can be used in tendon graft studies.[18–22]

Conclusion

The PT weave and STS techniques were created to aid patients towards early mobilization which in turn, could lead to better and faster patient outcome. In conclusion, this study evaluated the biomechanical properties of the PT and STS techniques in terms of maximum load and load to failure. The PT weave remains the gold standard for tendon grafting due to its strength and reliability, the STS technique showed encouraging results in terms of load to failure while also making up for some of the drawbacks of the PT weave such as its bulkiness and complexity.

The analysis found no significant difference in maximum load between the two techniques which could be attributed to differences in study protocols. However, there was evidence supporting that STS may have higher load to failure. Tendon failure locations further supported this, with PT failing linked to slipped sutures or donor tendon pull-through while STS failed because of longitudinal shear of the tendon.

Despite these findings for the STS technique, there were limitations such as the in vitro nature of the studies, variations with protocol and source of tendon materials. The sample size and potential heterogeneity in studies were also factors that could impact results. Further research should entail the use of in vivo models and more standardized methods to offer a clearer comparison between the two techniques.