➢ There is currently no cure for Dupuytren disease. Therapeutic interventions can treat the symptoms of the disease but do not prevent disease recurrence.
➢ Surgical treatment, needle aponeurotomy, and collagenase injections are effective at correcting initial Dupuytren deformities.
➢ Proximal interphalangeal joint contractures are more resistant to treatment than contractures at other joints and are prone to recurrent deformity.
➢ Collagenase clostridium histolyticum has demonstrated good clinical efficacy, is associated with a favorable side-effect profile, and is a viable alternative to surgical treatment for both primary and recurrent disease.
Dupuytren disease results from a pathologic proliferation of the palmar fascia, leading to the formation of constrictive fibrous cords and nodules that can severely impair hand function. Over the past decade, the treatment of Dupuytren disease has undergone a rapid evolution. Minimally invasive treatment options have emerged as viable alternatives to traditional palmar fasciectomy. This review will summarize the recent advances in clinical evaluation and treatment of Dupuytren disease.
Epidemiology and Etiology
The incidence of Dupuytren disease varies among geographic regions, with reported prevalence rates ranging from 0.2% to 56%1. The disease shows a predilection for individuals of northern European descent. It most commonly arises in middle-age patients between the ages of forty and seventy, with men being more commonly affected than women2,3. The sex discrepancy diminishes with increasing age, indicating a possible role for sex hormones as modulators of the disease process4,5. Diabetes, tobacco use, alcohol consumption, hypercholesterolemia, hyperlipidemia, human immunodeficiency virus (HIV) infection, epilepsy, trauma, carpal tunnel, frozen shoulder, and manual labor have all been suggested to have an association with the development of Dupuytren disease6-11. However, large-scale epidemiologic studies have been unable to demonstrate a definitive link between Dupuytren disease and each of these implicated factors12-14.
Dupuytren disease does show familial clustering, which suggests an underlying genetic mode of inheritance. Alterations in Wnt and transforming growth factor beta (TGFβ) signaling, mitochondrial mutations, chromosomal structural abnormalities, inflammatory pathways, and many other biological factors have been implicated. The largest genome-wide association study to date compared 960 Dutch patients who had Dupuytren disease with 3117 control patients. The investigators were able to show a linkage to nine genetic loci, six of which are known to involve Wnt signaling15. Proteomic analyses and RNA tissue profiling studies have provided some support for these observations by demonstrating links to Wnt and TGFβ signaling16-18. Family studies have shown conflicting results; a heteroplasmic mitochondrial mutation in a seven-generation family was found to be linked to the development of Dupuytren disease19. In contrast, a five-generation Swedish family did not show a mitochondrial inheritance pattern, but instead showed linkage to a region on chromosome 16q containing genes for various matrix metalloproteases as well as the NOD/CARD genes, which are associated with inflammatory response to infection20. The lack of a consensus among molecular studies underscores the complexity of what is likely a multifactorial disorder arising from a variety of different genetic alterations.
The palmar fascia is a thick fibrous matrix composed of longitudinally and transversely oriented fibers. It supports the underlying flexor tendons, metacarpals, and neurovascular structures of the hand while also serving as an anchor for the overlying skin. The terms cord and band were first proposed by Luck and are used to differentiate diseased and nondiseased palmar fascia, respectively21. Multiple cords can form in patients, with the most common being the pretendinous, central, spiral, natatory, lateral, abductor digiti minimi, retrovascular, and commissural cords22.
The pretendinous cords extend from the palmar fascia into the digits centrally, in line with the flexor tendon. Distally the cords insert into the skin and flexor tendon sheath just distal to the metacarpophalangeal (MCP) joint. Fibers emanate from the pretendinous cords and descend vertically along either side of the MCP joint and are contiguous with the vertical septa of Legueu and Juvara. The pretendinous cords lie central and superficial to the neurovascular structures and therefore do not cause displacement of those structures (Fig. 1). Contraction of the cord can produce flexion contractures at the MCP joint.
A distal extension of the pretendinous cord that crosses the proximal interphalangeal (PIP) joint and inserts into the volar skin at the level of the middle phalanx is known as the central cord. The central cord does not have a corresponding normal fascial precursor, and its contraction can cause flexion deformities at the PIP joints. As with the pretendinous cords, the central cord remains midline and does not alter the neurovascular anatomy.
The spiral cord originates just proximal to the MCP joint and can arise from the abductor digit minimi, intrinsic tendons, proximal phalanx, or pretendinous cord. The fibers start volar and central to the neurovascular bundle, coursing along its inner aspect and becoming dorsal and peripheral as they merge with the lateral digital sheet distal to the MCP joint. After contributing fibers to the lateral digital sheet, the spiral cord courses centrally, traversing around the volar aspect of the neurovascular bundle inserting into the flexor tendon sheath near the PIP joint. The pretendinous band, spiral band, lateral digital sheath, vertical band, and Grayson ligament all contribute to the spiral cord. As the spiral cord contracts, it pulls the neurovascular bundle proximal, central, and superficial (Fig. 2). Great care, therefore, must be taken during surgical treatment to avoid iatrogenic injury. Contracture of the spiral cord can produce severe flexion deformities at the PIP joint.
The lateral cord develops from the lateral digital sheet and runs along the sides of the affected digit, superficial to the neurovascular bundle. Proximally, the lateral cord originates from the interdigital web space arising from the natatory and spiral ligaments and attaches at its distal extent to the overlying skin and to the flexor tendon sheath through the Grayson ligament. Contracture of the lateral cord causes flexion contractures at the PIP joint and, less frequently, contractures at the distal interphalangeal (DIP) joint. The fibers of the lateral cord run parallel to the neurovascular bundle and rarely cause substantial displacement.
The abductor digiti minimi cord arises from the abductor digiti minimi muscle and emits fibers that can behave like a spiral or lateral cord (Fig. 3). Although this cord lies superficial to the neurovascular bundle, it can cause significant alterations in the anatomy. It is often close to the dorsal ulnar sensory nerve, and care must be taken during surgical dissection. The abductor digiti minimi cord can cause contractures at both the PIP and DIP joints.
The natatory cords are the only cords with transversely oriented fibers that are consistently affected in Dupuytren disease. They span the distal web spaces, limiting finger abduction. The natatory cords lie superficial to the neurovascular bundle and contribute fibers to the spiral and lateral cords. Natatory cords that form in the first web space are referred to as commissural cords; they can be quite debilitating as a result of their effects on thumb motion.
The retrovascular cord, as its name implies, lies deep (dorsal) to the neurovascular bundle; its fibers often merge with the lateral cord and can cause contractures at the DIP joints.
Patients with early Dupuytren disease most commonly present with an enlarging nodule in the palm of the hand. Nodules are usually painless, but can occasionally present with pain as they enlarge. When present, pain is thought to develop from local nerve irritation caused by proliferating fibroblasts23. Pain typically resolves without clinical intervention despite persistence and enlargement of the nodules. As time passes, nodules can expand and form cords that cause flexion contractures of the finger joints. The ring and small fingers (ulnar side) are the most commonly diseased digits, and the index finger is the least commonly affected24.
The clinical evaluation should include a thorough neurovascular examination and an assessment of range of motion for each finger, including the thumb. A deliberate attempt should be made to palpate pretendinous, central, spiral, natatory, lateral, abductor digiti minimi, retrovascular, and commissural cords. Patients may also have fibromatosis involving other areas of the hand and body, including the knuckle pads (Garrod disease) (Fig. 4), the penis (Peyronie disease), and the plantar fascia (Ledderhose disease). Bilateral disease, early age at onset, and systemic fibromatosis can be a sign of a more aggressive form of disease known as Dupuytren diathesis25.
The differential diagnosis for Dupuytren disease includes ganglia, inclusion cysts, stenosing tenosynovitis, callus formation, soft-tissue neoplasms, and rheumatoid nodules. In patients with joint contractures, camptodactyly, traumatic scars, neuropathy, and intrinsic muscle weakness, the differential diagnosis should also include joint disease, Volkmann ischemic contracture, PIP capsular contracture, spastic contracture, locked chronic neglected stenosing tenosynovitis, or contracture following cerebral vascular insult.
There is currently no cure for Dupuytren disease, and symptomatic recurrence following treatment is common. Despite the challenges, there are a variety of treatment strategies that can be used to alleviate symptoms and maximize hand function. Choosing an appropriate treatment plan involves a thorough discussion with the patient. Functional status, occupation, and personal preferences should all weigh heavily in the decision-making process. Over the past decade, a variety of new treatments have emerged, including needle aponeurotomy and collagenase injections. The indications for these treatments are constantly evolving as new data from ongoing clinical trials become available.
Physical therapy and splinting have not shown efficacy in reversing or preventing contractures associated with Dupuytren disease26. Corticosteroid injections can alleviate pain symptoms associated with symptomatic nodules and, in some patients, can produce partial resolution of nodules. In a series reported by Ketchum and Donahue, corticosteroid injections were able to reduce the size of symptomatic nodules in seventy-three (97%) of seventy-five hands. However, most patients had incomplete resolution of nodules and 50% of the patients developed recurrent disease within three years27.
Surgical treatment has traditionally been the treatment modality of choice for Dupuytren disease. Popular techniques include palmar fasciectomy, limited aponeurotomy, dermofasciectomy, and the McCash open palm technique. Surgical techniques permit the removal of diseased tissue and give excellent visualization of the neurovascular structures. Potential advantages include a greater degree of deformity correction, a more durable treatment response, and a lower incidence of iatrogenic injury to the underlying tendons and neurovascular structures.
A variety of skin incisions have been described for the treatment of Dupuytren disease28,29. Palmar fasciectomy is usually performed with use of a Bruner incision closed by a V-Y plasty (Fig. 5) or a longitudinal incision closed by a z-plasty. Citron and Nunez prospectively compared these two incisions and found no significant differences with regard to complication rate or disease recurrence30.
There are numerous studies reporting on the long-term follow-up results after palmar fasciectomy. Mavrogenis et al. retrospectively reviewed 196 patients with six-year follow-up after palmar fasciectomy. There was excellent correction of the initial deformity; however, all of the patients subsequently developed a PIP joint contracture of at least 20° and almost half of the patients subsequently developed an MCP joint contracture of at least 30°. There was a 5% incidence of neurovascular compromise, a 10% incidence of complex regional pain syndrome, and a 15% incidence of wound-healing complications31. In another series, Leclercq and Tubiana studied the ten-year follow-up data for fifty patients treated with palmar fasciectomy and reported similar findings. All of the patients were able to achieve full extension immediately after surgery, but 66% of the patients developed recurrent disease, with 24% lacking at least 45° of extension. Sixteen percent of the patients in their series required additional surgery to treat recurrent deformity32. These studies show that palmar fasciectomy is usually successful in correcting the initial deformity but that recurrence is common and the surgical complication rate is meaningful. PIP joint contractures appear to be more difficult to treat and are associated with a higher rate of disease recurrence.
A variety of factors have been associated with disease recurrence after palmar fasciectomy, including an age younger than forty-five years, a preoperative PIP joint flexion contracture >60°, incomplete surgical correction of the initial deformity, revision surgery, and poor compliance with postoperative regimens33,34. A meta-analysis of forty-one articles published over the last twenty years showed that the reported complication rate varied from 3.6% to 39.1% among studies. Wound-healing complications were the most common (22.9%), followed by complex regional pain syndrome (5.5%), digital nerve injury (3.4%), infection (2.4%), hematoma (2.1%), and arterial injury (2.0%). Revision surgery was associated with a tenfold greater likelihood of neurovascular injury35.
Dermofasciectomy with full-thickness skin grafting is a surgical alternative that is advocated by some as a means to lower the rate of recurrent disease. In a retrospective series reported by Ketchum and Hixson, there was no evidence of disease recurrence beneath skin-grafted areas at a mean follow-up of 3.9 years in sixty-eight patients treated with dermofasciectomy36. Armstrong et al. reported a favorable recurrence rate of 8.4% (twelve of 143 rays) at a mean follow-up of 5.8 years following dermofasciectomy37. Despite these outcomes, prospective clinical trials have not been able to demonstrate a superiority of dermofasciectomy over palmar fasciectomy. Ullah et al. carried out a prospective randomized controlled trial comparing fasciectomy with dermofasciectomy, which demonstrated an overall recurrence rate of 12.2% (eleven of ninety digits), which was not a significant difference between the two surgical groups38. Dermofasciectomy is currently not as popular as palmar fasciectomy, but it may have a role in treating patients with aggressive fibromatosis who are at high risk for early recurrence of disease.
The open palm technique, which was popularized by McCash, is a surgical alternative that seeks to minimize wound-healing complications and disease recurrence. The technique is performed by making multiple transverse incisions along the diseased fascia, leaving an intervening skin bridge between each incision39. Shaw et al. retrospectively reviewed thirty-nine digits treated with the open palm technique. In their series there was excellent correction of the initial deformity, with recurrence and complication rates that were comparable with those associated with open palmar fasciectomy. The potential advantages of this approach include lower risk of skin necrosis or wound-healing complications and better postoperative pain control. The disadvantages include inferior visualization of the neurovascular structures as compared with the visualization obtained during other surgical techniques, and prolonged wound-healing when performed without primary wound closure40.
Limited aponeurotomy is an alternative surgical technique that can be used to reduce the amount of time that a patient is under anesthesia. It is generally reserved for debilitated patients with poorly defined cords that are difficult to treat with either needle aponeurotomy or collagenase injections.
Needle aponeurotomy is a minimally invasive procedure that can be performed in the office setting under local anesthetic (Fig. 6). According to Eaton, four requirements should be met in order to perform needle aponeurotomy: (1) contracture due to a (2) palpable cord lying beneath (3) redundant skin in a (4) cooperative patient41. The advantages of needle aponeurotomy include a faster recovery time, fewer wound-healing complications, and a lower direct cost than either an operation or collagenase injection. Disadvantages include a potential increase in the rate of iatrogenic injury to underlying tendons and neurovascular structures, limited efficacy in treating severe or recurrent disease, a higher recurrence rate, and a steep learning curve.
Foucher et al. retrospectively reviewed the outcomes of needle aponeurotomy performed on 311 fingers in 211 patients. In their series, MCP and PIP joint contractures were corrected in 79% and 65% of the patients, respectively. The overall reoperation rate was 24% at a mean follow-up of 3.2 years, but was notably 59% for the first 100 operations performed, emphasizing the learning curve required to master this procedure. There were no tendon injuries or wound-healing complications, but there was one digital nerve injury42. van Rijssen and Werker performed a prospective study comparing needle aponeurotomy with open fasciectomy in 115 hands. They showed that fasciectomy produced a 79% improvement in extension versus a 62% improvement for needle aponeurotomy six weeks after the initial procedure. Despite a worse functional outcome, patients in the needle aponeurotomy group had less pain and greater satisfaction. At a follow-up of five years, there was an 85% recurrence rate in the needle aponeurotomy group versus a 21% recurrence rate in the fasciectomy group. In contrast to the six-week follow-up data, the five-year follow-up data showed a higher level of patient satisfaction after fasciectomy. The authors noted that recurrence was less likely in elderly patients and suggested that needle aponeurotomy be considered for elderly individuals and those willing to accept the risk of early recurrence43.
Additional studies suggest that more aggressive attempts at needle aponeurotomy have the potential to correct flexion deformities with a higher success rate than that associated with open surgical techniques. In a retrospective series by Pess et al., 1013 digits were treated with needle aponeurotomy for joint contracture of 20° or greater at the MCP or PIP joints. Joint contractures at the MCP joints showed an average deformity correction of 99% and PIP joint contractures showed an average correction of 89% immediately following needle aponeurotomy. At the time of final follow-up, 72% of the initial correction was maintained at the MCP joint, while only 31% of the joint correction was maintained at the PIP joint. The primary complication was skin tearing, which occurred in 3.4% of the patients44.
For more advanced disease, multiple needle aponeurotomy (more than four needling attempts per finger) may be needed to correct a flexion contracture. Multi-needle aponeurotomy for advanced Dupuytren disease was a prospective study that established the safety and efficacy of performing multiple needle aponeurotomy for advanced Dupuytren disease. Ninety-nine rays in thirty patients were treated, with the treatment resulting in excellent correction of the presenting flexion deformity. The correction was well maintained at a follow-up time of six-months. There was 100% patient satisfaction at the one month follow-up and 95% patient satisfaction at the six-month follow-up45.
Collagenase clostridium histolyticum, which was approved by the U.S. Food and Drug Administration in February of 2010, works by enzymatically digesting diseased cords until they can be physically broken by manipulation. It is a mixture containing class-I and class-II clostridial collagenase, which act synergistically to degrade constrictive bands of tissue more rapidly than human collagenases do.
Collagenase is administered by direct injection into the diseased cord. If the needle is inserted too deeply, there is a risk that the collagenase could damage the underlying flexor tendons46. Eighteen to twenty-four hours after the injection is performed, the affected finger is extended in a controlled manner, effectively “breaking” the diseased cord. After releasing the diseased cord it is important to check active finger flexion to assess the flexor tendon status. Complications associated with collagenase injection include allergic reactions, skin tears, tendon rupture, injury to nerves and arteries, bruising, pain at the injection site, regional lymphadenopathy, complex regional pain syndrome, and skin discoloration (Fig. 7).
JOINT-I (United States) and JOINT-II (Australia and Europe) were the first phase-III clinical trials to evaluate the safety and efficacy of collagenase clostridium histolyticum. Both trials were nine-month open label studies evaluating 879 joints in 587 patients. Each study participant was required to have at least one finger affected by a Dupuytren contracture measuring more than 20°. Flexion contractures were reduced to 5° or less for 57.0% of the joints treated, and MCP joints had a better overall treatment response than did PIP joints. There were no major adverse events reported in either trial47. One year post-marketing safety data on collagenase continues to support the safety profiles observed in JOINT-I and JOINT-II. One hundred and fifteen adverse events were observed after 5400 injections, and the majority of them were considered minor (skin tears, edema, bruising). Serious adverse events occurred in 0.6% of the patients and included two flexor tendon ruptures and one flexor pulley injury48.
After the safety and efficacy of collagenase clostridium histolyticum was established in JOINT I and JOINT II, the Collagenase Option for Reduction of Dupuytren trials I and II (CORD I and CORD II) were carried out to evaluate the outcomes of collagenase therapy. CORD I and CORD II were both twelve-month placebo-blinded prospective studies. The primary end point was correction of a flexion contracture to within 5° of full extension within thirty days after the last collagenase injection. Three hundred and eight patients were enrolled in CORD I, and sixty-six patients were enrolled in CORD II. When compared with placebo, collagenase injections showed a success rate of 64% compared with 6.8% for the placebo group in CORD I and a success rate of 44% compared with 5% for the placebo group in CORD II49-52.
Eight-year follow-up data from these early stage trials was reported by Watt et al. Eight of twenty-three patients from their institution completed the full eight-year follow-up. Six of the patients had injections for MCP joint contractures, and two had injections for PIP joint contractures. Four of the six patients in the MCP group had recurrent disease, whereas both patients with PIP joint contractures developed recurrence. Despite the high recurrence rate, patient satisfaction was high—88% of the patients indicated that they would have collagenase injections again for treatment of recurrent disease53.
Long-term outcomes of collagenase therapy are currently being evaluated in the CORD Long-term Evaluation of Safety Study (CORDLESS). This study includes 1080 joints in 643 patients. The recently published three-year follow-up data shows a 56% recurrence of flexion contracture at the PIP joint and a 27% recurrence rate at the MCP joint. Seven percent of the joints that were successfully treated required repeat collagenase injections or surgical treatment. Three hundred and one of the joints evaluated in this trial were considered to have measurable improvement but did not meet criteria to be considered a successfully treated joint. Of this group, 50% developed clinically important recurrent deformity, with 9% requiring surgical correction. These data demonstrate a higher recurrence rate for patients with only partial deformity correction (50%) versus those with deformity correction to within 5° of full extension (35%). There were 370 adverse events reported in 193 of the 643 patients; 4.8% were considered severe, while the majority were considered mild or moderate. At least 82% of the patients developed autoantibodies after receiving one collagenase injection; this percentage increased to 96% after two injections, although the clinical significance of this is currently unclear46,54.
Bainbridge et al. retrospectively reviewed 1082 patients from twelve clinical trials to evaluate the efficacy of collagenase in patients who had undergone previous surgery for Dupuytren disease. There was no significant difference between patients who did or did not have previous surgery with regard to the amount of deformity correction that was obtained at the MCP and PIP joints and with regard to overall satisfaction with treatment. However, there was a significant increase in hemorrhage at the injection site (p = 0.05), blood blister (p = 0.01), and axillary pain (p = 0.01) for patients who had undergone previous surgery. In total, there were ten serious adverse events reported, while the majority of the adverse events were considered minor and resolved without intervention55. There have been reports of skin graft loss following collagenase injections, leading several authors to urge caution when using collagenase at or near previous skin graft sites, regardless of the age of the graft56,57.
Raven et al. evaluated the safety and efficacy of collagenase in different population subgroups by combining data from three randomized, placebo-controlled, double-blinded trials. There was not a significant difference in clinical success and adverse events for patient subgroups with regard to age, diabetes status, and sex58. These results support the use of collagenase in high-risk population subgroups.
Chen et al. carried out a cost-utility analysis comparing palmar fasciectomy, needle aponeurotomy, and collagenase injection. Treatment was considered cost-effective on the basis of the willingness to pay $50,000 per quality-adjusted life year gained. The authors concluded that open fasciectomy was not cost-effective, needle aponeurotomy was cost-effective if the success rate was near 100%, and collagenase injection was cost-effective if it was priced at $945 or less per injection59,60. Palmar fasciectomy, needle aponeurotomy, and collagenase treatments all show good short-term efficacy in correcting flexion deformities in patients with Dupuytren disease; however, long-term outcome data evaluating disease recurrence after each of these treatment options are needed to determine which treatment options are truly the most cost-effective. With growing concerns over the rising costs of health care, future studies examining the cost-to-benefit ratio associated with the use of therapeutics for Dupuytren disease will likely play a major role in guiding clinical decision making.
Surgical treatment, needle aponeurotomy, and collagenase injection are viable treatment options for Dupuytren disease. They all demonstrate excellent short-term efficacy in correcting joint deformity with acceptable complication rates. Disease recurrence requiring repeat interventions continues to be the main challenge. As long-term outcome data become available, our understanding of the safety and durability of the various treatment options will help guide clinicians and patients in choosing the most appropriate treatment modalities.
Source of Funding: No outside funds were used for this study.
Investigation performed at the Department of Orthopedic Surgery, Mayo Clinic, Rochester, Minnesota
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
- Copyright © 2014 by The Journal of Bone and Joint Surgery, Incorporated