Introduction
Dupuytren's disease (DD) is a benign and progressive fibroproliferative disorder of
the palmar fascia, characterized by the formation of nodules and cords that may evolve
into flexion contractures, most frequently affecting the metacarpophalangeal (MCP)
and proximal interphalangeal (PIP) joints.[1]
[2] The condition results from the abnormal proliferation and differentiation of fibroblasts
into contractile myofibroblasts, leading to thickening, shortening, and loss of digits'
extension.[1]
Although its etiology is multifactorial, with strong genetic influence, several modifiable
and nonmodifiable risk factors have been consistently associated with the disease,
including increased age, male sex, alcohol consumption, smoking, diabetes mellitus,
epilepsy, liver disease, and occupations involving manual labor or vibratory exposure.[3] The estimated prevalence in the general population ranges from 4 to 6%, with higher
rates observed in individuals of Northern European ancestry and older age groups.[2]
[4] In advanced stages, the disease can significantly impair hand function and quality
of life.
Treatment options for DD have evolved over centuries, with surgical interventions
such as fasciotomy and fasciectomy being the primary approaches since the 17th century.[5] In recent years, less invasive treatments have been increasingly explored, particularly
intralesional ones that aim to modify disease progression in early stages or offer
alternatives to surgery. Injectable collagenase Clostridium histolyticum (CCH), which enzymatically disrupts the collagen cords, has demonstrated short-term
efficacy in reducing contractures, especially at the MCP joint, and became widely
used after regulatory approval, although its availability has varied by region.[6] Other agents such as corticosteroids, injected into early nodules, have been associated
with partial regression or disease stabilization,[7] while biologic therapies targeting molecular pathways, including anti-tumor necrosis
factor (TNF) agents, are under investigation with promising early-phase results.[8]
Among surgical techniques, percutaneous needle fasciotomy (PNF) represents a minimally
invasive option associated with rapid recovery and low morbidity. However, it carries
a high rate of recurrence, particularly in patients with aggressive disease. Limited
fasciectomy remains the standard treatment for more advanced contractures, offering
greater durability at the cost of longer recovery and higher complication rates.[5]
[8] In recurrent or severe cases, dermofasciectomy may be indicated, although it involves
skin grafting and presents higher surgical complexity and morbidity. The choice of
intervention depends on multiple factors including severity, joint involvement, comorbidities,
recurrence risk, and surgeons' experience.[9]
Despite the availability of various treatments for DD, there is a notable gap in the
literature regarding comprehensive comparisons of these interventions. The lack of
robust comparative syntheses limits the ability of clinicians to identify which interventions
offer the best balance of effectiveness, safety, and long-term outcomes, particularly
with respect to functional improvement, adverse effects, and recurrence rates.
Previous systematic reviews have typically focused on individual treatment modalities—such
as collagenase, needle fasciotomy, or limited fasciectomy—excluding other relevant
surgical techniques and injection therapies, such as corticosteroid or experimental
biological agents.[10]
[11]
[12]
[13]
[14] Only one review conducted a network meta-analysis (NMA) and, even then, the comparison
was restricted to three interventions.[15] Only one Cochrane review, published in 2015, attempted broader comparisons but included
only surgical techniques and did not incorporate newer modalities.[16]
Several recent reviews have also presented important methodological limitations that
hinder their applicability in clinical decision-making. Many restricted inclusion
to studies published in English or sourced only from traditional databases, potentially
overlooking relevant data from other regions or languages.[10]
[11]
[15] Additionally, the selection of patient-reported outcome measures (PROMs) has been
inconsistent, often lacking a clear hierarchy or justification.[11]
[12]
[13] Several recent reviews did not include NMA, thereby restricting their capacity to
compare interventions beyond direct pairwise analyses.[10]
[11]
[12]
[13]
[14]
Finally, most of the existing reviews are now methodologically outdated, having completed
their literature searches more than 5 years ago, prior to the publication of several
key randomized controlled trials.
Collectively, these limitations reinforce the need for an updated, comprehensive,
and methodologically rigorous systematic review to support evidence-based clinical
decision-making. This protocol outlines the methodology for a systematic review and
NMA aiming to compare the effectiveness and safety of surgical and injection-based
interventions for the treatment of DD.
Materials and Methods
The present review was conducted according to the Preferred Reporting Items for Systematic
Reviews and Meta-Analyses for Network Meta-Analyses (PRISMA-NMA) guidelines.[17] This study has been Registered on The International Prospective Register of Systematic
Reviews (PROSPERO) database, under the number CRD42023439436. The Preferred Reporting
Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) checklist was
used for the development of this protocol.[18] Any substantial amendments will be documented and updated in the PROSPERO registry.
Criteria for Considering Studies
Study Population and Study Design
The proposed review must include randomized controlled trials (RCTs) evaluating surgical
and injection-based interventions for DD. Eligible studies must include participants
aged 18-years or older, diagnosed with DD, and treated with either surgical or injection
interventions. To ensure the assessment of long-term outcomes, only trials with a
follow-up period longer than 3 months should be included. No restrictions must be
applied regarding publication status or language, allowing the inclusion of both published
and unpublished studies.
Definitions and Outcome Measures
The studies must include a comparison arm that includes infiltration and/or surgical
treatment. Studies comparing the following interventions were eligible for inclusion
in this review:
-
Surgical interventions: total or partial fasciectomy; percutaneous fasciotomy, among
others.
-
Infiltration interventions: any treatments using active substances such as CCH or
corticosteroids, among others.
The primary outcomes of the proposed review will focus on functional improvement and
adverse effects. Functional improvement can be assessed using patient-reported scales,
following a predefined hierarchy of instruments, from more to less disease-specific
scales ([Table 1]).[19]
[20]
[21]
[22]
[23]
[24]
Table 1
Hierarchy of PROMs for assessing functional improvement in DD
|
Rank
|
Instrument
|
|
|
1
|
SDSS[19]
|
|
|
2
|
DIF-CHUM[20]
|
|
|
3
|
URAM[21]
|
|
|
4
|
DASH and/or QuickDASH[22]
|
|
|
5
|
MHQ and/or briefMHQ[23]
|
|
|
6
|
CHFS[24]
|
|
Abbreviations: CHFS, Cochin Rheumatoid Arthritis Hand Function Scale; DASH, Disability of the Arm,
Shoulder, and Hand Questionnaire; DD, Dupuytren's disease; DIF-CHUM, Dupuytren's Contracture
Impact on Function - Centre Hospitalier de l'Université de Montréal; MHQ, Michigan
Hand Questionnaire; PROM, patient-reported outcome measure; SDSS, Southampton Dupuytren's
Scoring Scheme; URAM, Unité Rhumatologique des Affections de la Main.
Adverse effects should be categorized into minor and major complications and ranked
by order of interest for analysis. Minor complications include superficial infection,
wound dehiscence, delayed healing, and hematoma. Major complications consist of deep
infection, the need for reoperation, skin coverage, ischemia, necrosis, nerve injury,
tendon rupture, intensive care unit admission, and death. The following complications
will be prioritized in the analysis, in descending order of interest: nerve injury,
delayed healing, need for reoperation, superficial infection, and deep infection.
Secondary outcomes will include range of motion, measured through contracture angle,
total range of motion, or Tubiana's stage. Pain will be assessed using the visual
analog scale (VAS). Recurrence will be defined as more than 20 degrees of contracture
in any treated joint 1-year after treatment, compared to 6 weeks posttreatment, with
each treated joint evaluated individually.[25]
Search Methods for Identification of Studies
Relevant studies should be identified by searching the electronic databases and clinical
trial registries: Cochrane Library, Medline, Embase, Lilacs, Ibecs, ClinicalTrials.gov,
and the World Health Organization International Clinical Trials Registry Platform.
These searches will be conducted without language or publication date restrictions.
The full search strategy for each database is provided in [Appendix 1]. Additionally, manual searches should include reviewing reference lists from relevant
conference abstracts and proceedings, and content experts and authors must be contacted
to identify unpublished or ongoing studies.
Data Collection and Analysis
Study Selection
The screening process should be conducted using the Rayyan (Qatar Foundation) software.
Two authors must independently screen titles and abstracts to identify potentially
eligible studies based on predefined eligibility criteria. Full-text articles should
be retrieved for studies included in the initial screening, followed by assessment
of their eligibility for final inclusion. In the event of disagreements, a third reviewer
should be consulted to resolve conflicts. The selection process must be documented
in a PRISMA flowchart, detailing reasons for excluding studies.
Analysis and Data Synthesis
Data extraction and management should be conducted independently by two authors using
a standardized form in Microsoft Excel (Microsoft Corp.). Extracted data must include
study methods (design, definition), identification data (sponsorship, country, author
details), participant characteristics (number randomized, analyzed, and lost to follow-up),
baseline data (age, sex, disease severity, body mass index [BMI], disease duration),
interventions (description, group sizes), outcome measures (type, scale, unit), study
design characteristics, and risk of bias assessment. Tables summarizing study characteristics
and risk of bias must be created. For continuous outcomes, means, standard deviations
(SD), and participant numbers should be extracted, with data transformation as needed.
For dichotomous outcomes, event rates and participant numbers will be extracted. Intention-to-treat
analysis should be applied whenever possible. Discrepancies must be resolved through
consensus, and final data must be exported to RStudio (Posit Software, PBC) for analysis.
For continuous outcomes, means, SDs, and participant numbers should be extracted.
If alternative measures of dispersion are reported (e.g., confidence intervals or
standard errors), SDs must be be calculated according to the Cochrane Handbook. Standardized
mean differences (SMD) must be used for outcomes reported on different scales, with
results expressed as 95% confidence intervals (CIs). For dichotomous outcomes, event
rates and participant numbers must be extracted to calculate odds ratios (ORs) with
95% CIs.
The unit of analysis considered is each participant. For studies reporting multiple
time points, data from the longest follow-up must be extracted. For missing data,
study authors should be contacted. If unavailable, the study must proceed with the
available data, and sensitivity analyses will assess the potential impact. Intention-to-treat
(ITT) analysis should be used whenever possible.
Heterogeneity must be be assessed based on participant characteristics, interventions,
and outcomes. The statistical results should be be evaluated using forest plots and
quantified with the I2 statistic, and a Chi-squared test (p < 0.1) to detect significance. Heterogeneity is to be classified as not important
(0–40%), moderate (30–60%), substantial (50–74%), or considerable (75–100%). Subgroup
analyses can be conducted to explore potential sources of heterogeneity.
Tables with a summary of findings should be prepared for all outcomes, using the most
appropriate reference comparator. Certainty of evidence will be assessed using the
Grading of Recommendations, Assessment, Development and Evaluation (GRADE) approach
by two authors. Certainty must be downgraded based on factors such as study limitations,
inconsistency, and publication bias, with results categorized as high, moderate, low,
or very low. Disagreements should be resolved by consensus, with justifications for
downgrades placed in the footnotes.
Subgroup analyses should investigate heterogeneity by age, comorbidities (e.g., diabetes,
epilepsy, alcoholism, smoking), disease duration, severity (Tubiana stage/total range
of motion), and surgical techniques (e.g., open palm, McCash, Bruner incision, skin
graft, z-plasty). Sensitivity analyses should be conducted to test the robustness
of findings, focusing on the inclusion of unpublished studies and those with a high
risk of bias.
Quality Assessment
The risk of bias for each included study must be independently assessed by two review
authors using version 2 of the Risk of Bias (RoB2, Cochrane Collaboration) tool, following
the recommendations in the Cochrane Handbook for Systematic Reviews of Interventions,
version 6.3. Any disagreements must be resolved through consensus. The assessment
should focus on bias across several domains: bias arising from the randomization process,
deviations from intended interventions, missing outcome data, measurement of outcomes,
and selection of reported results. For each domain, responses to signaling questions
will result in a judgment of “low risk of bias,” “some concerns,” or “high risk of
bias.” An overall assessment must be provided based on the RoB2 tool's algorithm.
Publication bias should be assessed using funnel plots and Egger's test, provided
there are at least 10 studies available for direct comparison. The NMA should be conducted
for both direct and indirect comparisons of interventions. Random-effects frequentist
NMAs and pairwise meta-analyses should be used to assess transitivity and inform the
strength of evidence. Coherence between direct and indirect evidence should be evaluated
using the split node method.
Treatment rankings should be calculated based on the P score, an analog to the surface
under the cumulative ranking (SUCRA), to determine the probability of each intervention
being the most effective. These values can be used to rank treatments according to
precision-weighted point estimates. All analyses must be performed using the netmeta
package, version 2.8-0, in R (R Foundation for Statistical Computing).
Ethics and Dissemination
The proposed systematic review follows the guidelines of the Brazilian National Health
Council. As the review involves synthesizing previously published and anonymized data,
it does not require formal review or approval from the Research Ethics Committee.
A Declaration of Responsibility confirming this exemption is available upon request.
The findings from this review will be disseminated through publication in peer-reviewed
journals and presentations at relevant scientific conferences. The data will be made
available upon reasonable request, following best practices in curation and deposition.
Discussion
As a progressive fibroproliferative condition, DD significantly impacts hand functionality
and quality of life.[5]
[26] Despite the availability of multiple treatments, such as percutaneous needle fasciotomy,
fasciectomy, and CCH injections, the outcomes remain inconsistent, and recurrence
rates are high.[6]
[27] The absence of standardized definitions for recurrence and the heterogeneity in
outcome measures complicate direct comparisons between existing studies.[28]
[29]
[30] This protocol aims to address these gaps by employing a systematic approach to synthesize
robust comparative evidence.
The present study adopts a PRISMA-based methodology to ensure transparency and reproducibility.[17] Key strengths include the application of NMA, which allows for both direct and indirect
comparisons among multiple interventions, and the integration of planned subgroup
analyses—considering variables such as patient age, comorbidities, and disease severity—to
improve clinical applicability. Nonetheless, the review is expected to face challenges
inherent to the literature, particularly the heterogeneity in outcome reporting and
variability in study quality.[13]
[28]
[30] This variability poses a potential source of bias, which could affect the robustness
of the planned analyses. These factors may introduce bias and affect the strength
of the resulting evidence.
The need for a new systematic review on this topic is reinforced by the methodological
limitations consistently observed in recent reviews. Most available studies have evaluated
only two or three interventions, typically PNF, collagenase, and limited fasciectomy,
thus excluding relevant options of treatment. Additionally, outcome reporting has
been heterogeneous, with little to no use of structured PROM hierarchies. Most reviews
have also conducted their literature searches over 5 years ago, thus failing to capture
several high-quality randomized trials published more recently.[13]
[15]
[16] Collectively, these limitations highlight a gap in the literature that justifies
the implementation of a comprehensive, up-to-date, and methodologically robust systematic
review, as proposed in the present protocol.
The findings from the proposed review have the potential to inform treatment decisions
and improve the clinical management of DD. By addressing critical gaps, especially
the lack of standardized definitions and consistent outcome measures for functional
improvement and adverse effects, this study may contribute to more uniform and reliable
assessments in future research.
