CC-BY-NC-ND 4.0 · Rev Iberoam Cir Mano 2017; 45(02): 070-088
DOI: 10.1055/s-0037-1608749
Original Article | Artículo Original
Thieme Revinter Publicações Ltda Rio de Janeiro, Brazil

Efficacy of Collagenase Clostridium Histolyticum for Dupuytren Disease: A Systematic Review

Article in several languages: English | español
Rafael Sanjuan-Cerveró
1  Department of Orthopedics and Traumatology Surgery, Hospital de Dénia, Alicante, Spain
,
Pedro Vazquez-Ferreiro
2  Department of Ophthalmology, Hospital Virxe da Xunqueira, La Coruña, Spain
,
Diego Gomez-Herrero
3  Pharmacy Service, Hospital 9 de Octubre, Valencia, Spain
,
Francisco Javier Carrera-Hueso
4  Pharmacy Service, Hospital Dr. Moliner, Valencia, Spain
› Author Affiliations
Further Information

Address for correspondence

Rafael Sanjuan Cerveró, MS
Department of Orthopedics and Traumatology Surgery
Hospital de Dénia, Partida de Beniadla s/n, 03700, Dénia, Alicante
Spain   

Publication History

26 July 2017

19 September 2017

Publication Date:
22 November 2017 (eFirst)

 

Abstract

Introduction Collagenase Clostridium Histolyticum (CCH) has become a therapeutic alternative for Dupuytren disease. However, its efficacy in the medium to long term is unknown. The objective of our study is to carry out a systematic review of the studies conducted on the subject.

Material and Methods Systematic bibliographic search. Analysis depending on the time of progression, looking into 2 groups with the follow-up cut-off point of 1 year. Analysis of the number of patients who reached the primary endpoint, of the mean correction in degrees, and of the proportional correction of each joint.

Results The 50 selected clinical trials encompass a total of 4,622 patients (an average of 92.70). A total of 7,546 joints were treated with the mean being 148.15 joints per trial (3,925 metacarpophalangeal [MCP] and 2,350 proximal interphalangeal [PIP]). In less than one year of progression, the primary end point was reached in 48.9% of the joints (69.77% of the MCPs and 30.14% of the PIPs), the mean correction in degrees was 45.5 (standard deviation [SD]: 19.18) degrees; 40.8 degrees in the MCP (SD: 10.12) and 35.6 in the PIP (SD: 13.23), and the proportional correction of the joints was 72.9% (SD: 14.43) (83.9% for MCPs [SD: 12.58] and 64.2 for the PIPs [SD: 16.35]). In the follow-ups over 1 year, the primary end point was reached at a rate of 57.5% (68.9% of the MCPs and 43.3% of the PIPs), the mean correction in degrees was 37.6 degrees (SD: 10.93) (37.3 degrees in the MCPs [SD: 9.98] and 23.7 in the PIPs [SD: 16.33]) and the proportional correction of the joints was 87.3% (SD: 10.96) (90.3% for MCP [SD: 6.94] and 75% for PIP [SD: 13.54]).

Conclusions The results indicate a satisfactory response to CCH treatment maintained in the short and medium term. The recurrence rate is uncertain given the available data.


#

Introduction

The collagenase clostridium histolyticum (CCH) used for Dupuytren Disease (DD) is a mixture of two enzymes that degrade collagen types I and III found in the abnormal tissue that constitutes this fibromatosis. The first attempt to inject substances inside the cord related to DD was made by Bassot,[1] [2] in the 1960s. He coined the term “pharmacodynamical exeresis” and obtained relatively good results, published in 1969[3], looking for a proteolytic, anti-inflammatory and anesthetic effect. The term “enzymatic aponeurotomy” was adopted in 1971 by Hueston[3], who slightly modified the mixture for the injection. The treatment is identified as a valid alternative for patients who cannot be treated by the usual means. Using a similar technique, McCarthy obtained good results suggesting that the technique is an effective alternative as a substitute for fasciectomy.[2] Finally, the studies by Hurst and Badalamente[4] [5] established the efficacy of the treatment with an enzyme, the CCH, that allowed to break the cord in a local and minimally invasive manner.

Since the publication of the first clinical trial with CCH[6] and its marketing in the USA and Europe, this drug has become more important over time[7] [8] with regard to the treatment of DD. The development of the CORDLESS clinical trial,[9] [10] which involved the follow-up of patients enrolled in four previous clinical trials, has improved the level of understanding of progression in the short and medium terms. Likewise, these studies have been the basis of other publications regarding the analysis of subgroups[11] and partial results.[12]

Numerous clinical series and comparative studies have been published since then, providing an independent point of view regarding the clinical results obtained with the treatment. The variability among them is the norm rather than the exception, and the comparability between studies is a complex matter.[13] Although CCH is nowadays an alternative adopted by numerous hand surgeons in the treatment of DD, the rate of recurrence is unknown; thus, its status in the therapeutic arsenal compared with surgery is not clear. The scientific evidence at this time is also limited since no systematic reviews have been performed so far, and comparative studies with other techniques have been limited. The objective of our study is to carry out a systematic review covering all these studies, which evaluate independent studies and demonstrates the result of the CCH treatment since its commercialization aimed at assessing the efficacy of the treatment.


#

Material and Methods

A structured bibliographic search was performed in the PubMed, Google Scholar, Ovid and Web of Science databases with the following strategy: (Dupuytren disease [MeSH Terms]) AND (collagenase [MeSH Terms] AND clostridium histolyticum [MeSH Terms]). The search covered articles published from September 3, 2009 (date of publication of the CORD I study[6]) until June 15, 2017. A search of gray literature was performed in the databases of doctoral theses and at the US National institutes of Health Clinical Trials (www.clinicaltrials.gov), as well as through a manual review of the bibliography included in the articles.

We initially included all designs of cohort studies, clinical trials, case-control and case series published in English, Spanish, German, French and Italian; and those in which patient follow-up was specified for at least 30 days. Only studies that included patients diagnosed with Dupuytren contracture susceptible to surgical intervention with an initial contracture degree equal to or greater than 20°, and with at least one group of patients treated with CCH were considered. Reanalyzes of previous series, or those that did not provide data for the analysis, were excluded.

A structured form was used for the extraction and definitive collection of the data from the studies selected independently by two authors. The discrepancies that emerged upon comparing the results of both authors were resolved by a third author. To assess bias in the studies, we used the strategic orientation of business enterprises (STROBE) scale, with two researchers scoring each of the selected studies and using the mean of the two to assess their quality. Any disagreements were resolved through consensus with a third researcher.

The main variables collected from each study were the features and the design of the study, the interventions performed, the CCH doses utilized, the number of injections per patient, the extension time, and the criteria regarding severity and results, as well as the follow-up time. Finally, the funding from each study, if any, was collected. In clinical trials and comparative observational studies, only the clinical results of the group of patients treated with CCH were taken, and they were taken as a sample reference.

An analysis of the existing literature was conducted with regard to the results, taking into consideration two groups with a follow-up cut-off of 1 year. The clinical results were assessed based on three parameters: A) The number of patients who have reached the primary end point established in the CORD studies[6] [14] (final extension after treatment between 0–5°) stating the result in total absolute percentage of joints that have reached the objective; B) The assessment of the mean correction of each joint in degrees was evaluated, defined as the result of the degree of initial contracture minus the degree of final contracture; and C) Data regarding the correction ratio of the treated joint was also collected. If an article stated the results of two different modes, both have been included. An assessment of the recurrences over time has been made in the studies intended for that purpose,[9] [10] as well as in all those that cited them in their series. Finally, the concept of “non-effective” treatment in the series has been assessed. The missing data in the tables explain the lack of direct correlation between results (A + B≠C, as in one study overall results may have been given and not the results related to the metacarpophalangeal [MCP] and proximal interphalangeal [PIP] joints, for example).

The calculation of the results stated in the text has been performed as follows. The primary end point included in the CORD studies[6] [14] was established as the primary reference for the results. The secondary end point is similarly indicated in the results of the CORD studies,[6] [14] such as the patients who showed at least 50% of improvement since the initial contracture. In our analysis, the reflected number of these patients is that of those who have reached this point, while those included in the primary end point have been excluded, thus specifying the patients with improvement greater than 50%, but excluding those who have reached extensions between 0–5 degrees. The mean correction of the affected joints was obtained by acquiring direct data from the corresponding article, or the difference between the initial mean degrees of the middle and the endpoints. Finally, the proportion or percentage of the correction achieved was obtained in the same way, indicating the correction percentage for each of the treated joints.

The immediate measures to be implemented following treatment with CCH is a controversial issue. Therefore, in the articles analyzed, usage of orthosis and referrals to a physiotherapy protocol following CCH administration have been reviewed. In the final chapters, we analyzed both the recurrences and the treatment failures, commonly considered in the literature as “non-responders.”


#

Results

Search Results

Total 598 articles were obtained, of which only 240 studies have met the inclusion criteria. After removing duplicates, the elimination followed the exclusion criteria following the reading of the review summary, clinical cases, letters to the director, editorials and meeting summaries. Total 61 articles were obtained for analysis. These studies were reviewed through a complete reading of the article. Eleven studies were eliminated for a variety of reasons: 6 due to insufficient data on clinical outcomes (including cost studies), 3 due to cross-references to previous studies (subgroup analysis), and 2 due to exclusive references to results related to thumbs (excluded from the analysis of the CORD study[6] [14]). In the end, 50 publications were analyzed ([Fig. 1]). The interobserver assessment performed using the STROBE scale of the included articles has shown great homogeneity among the researchers (kappa > 0.85).

Zoom Image
Fig. 1 Flowchart of the selection of articles analyzed.

#

Results of the Clinical Studies

The main features of the selected studies are shown in [Table 1]. The monitoring of progression corresponding to the 3-year CORDLESS study has been included in the series,[9] taking into account that the patients associated with the CORD I, CORD II, POINT I and POINT II studies have been excluded for the calculation of demographic data.[6] [14] [15] The POINT X study was also excluded[16] because it was conducted in a subgroup of patients associated with the European POINT II study. The median follow-up of patients was 265 days (range 14–1,095; standard deviation [SD]: 233 days). The mean number of patients per study was 95 with a very variable range between 714[17] and 8, in the first independent series published in Germany[18] ([Table 2]).

Table 1

Article

Date

Type

Progress

Center

Country of origin

Study design

Follow-up

Severity

Dose

TRT Cycle

Measure

Extension

Hurst[6]

2009

R

P

M

USA

double blind randomized 2:1

90

CORD

S

1–3

(1.44)

ROM

24

Gilpin[14]

2010

R

P

M

Australia

double blind randomized 2:1 + Open

90

CORD

S

1–3

(1.5)

ROM

24

Spanholtz[18]

2011

CS

P

U

Germany

Case series. Follow-up

14

Tubiana

S

(1–3)

Ext def

24–30

Coleman[57]

2012

S

P

U

Australia

1st step, 1 dose; 2nd step, 2 doses on the same hand

270

CORD

S

1

ROM

24

Peimer[9]

2013

CS

P

M

Europe - Australia - USA

Follow-up CORD I + CORD II + JOINT I + JOINT II

1,095

CORD

S

(1–3)

ROM

Peimer[65]

2013

CS

R

M

USA

Retrospective on clinical histories

30

CORD

S

1.08

(1–3)

ROM

24

Vollbach[19]

2013

C

P

U

Germany

Comparison with fasciectomy

365

S

ROM

Skirven[32]

2013

C

P

U

USA

Comparison between IFP with physiotherapy and without it

30

CORD

S

1

Ext def

24

Marmol[43]

2013

CS

P

U

Spain

Follow-up

30

Tubiana

S

1

Ext def

24

McMahon[35]

2013

CS

P

M

USA

Retrospective on clinical histories

450

S

Ext def

24

Alberton[44]

2013

CS

P

U

Italy

Limited to one joint

180

Tubiana

S

1

Ext def

24

Martin-Ferrero[45]

2013

CS

R

U

Spain

Limited to 1 joint

365

Tubiana

S

1

Ext def

24

Nydick[24]

2013

C

P

U

USA

Comparison with aponeurotomy with needle

180

S

1–3

Ext def

24

Sanjuan[28]

2013

C

R

U

Spain

Cost study

180

Tubiana

S

1

Ext def

24

Witthaut[15]

2013

CS

P

M

USA - Australia - Europe

Open

270

CORD

S

1–3

(1.4)

ROM

24

Binter[51]

2014

CS

R

U

Austria

Retrospective review

365

CORD

S

1

ROM

24

Mickelson[40]

2014

R

P

U

USA

Differences between extension time

30

S

1

Ext def

1–7

Considine[52]

2014

CS

P

U

Ireland

Follow-up

45

S

1

Ext def

24

Garcia-Olea[64]

2014

CS

P

M

Spain

Follow-up

365

S

Ext def

Muppavarapu[20]

2014

C

R

U

USA

Comparison with fasciectomy

426

CORD

S

1.08

(1–3)

ROM

24–48

Sood[30]

2014

CS

R

U

USA

Selection of ”veteran” patients

369

Own criteria

S

1.6

(1–3)

Ext def

24

Coleman[46]

2014

C

R

M

USA - Australia

Two doses on the same hand at the same time

60

S

1

ROM

24

Leclere[33]

2014

CS

R

U

Switzerland

Ultrasound-assisted injection

300

S

1.17

(1–3)

Ext def

24

Povlsen[21]

2014

C

P

U

UK

Comparison with fasciectomy

S

1

ROM

72

Manning[41]

2014

CS

R

U

UK

Comparison of the extension times after the injection

98

S

1

Ext def

48

Warwick[38]

2014

CS

P

M

Europe

Functional recovery and health resources

180

AV

1–2

Ext def

24–48 x

Verheyden[39]

2014

CS

R

U

USA

Revision and dose increase

60

AV

1

Ext def

24

Perez-Giner[53]

2015

CS

P

U

Spain

Follow-up

90

CORD

S

1.19

(1–3)

Ext def

24

Considine[52]

2015

CS

P

U

Ireland

Limited to one dose

45

CORD

S

1

Ext def

48

Zhou[22]

2015

C

R

M

Netherlands

Comparison with fasciectomy

84

CORD

S

(1–3)

Ext def

24–72

Kaplan[42]

2015

CS

P

M

USA

Randomized. Extension at different times

90

CORD

S

1

Ext def

24–96

Gaston[17]

2015

CS

P

M

USA - Australia - New Zealand - Europe

Two doses on the same hand

60

CORD

DB

1

ROM

24–48–72

Atroshi[47]

2015

CS

P

U

Sweden

Follow-up

21

CORD

AV

1–2

Ext def

24–48 x

Tay[23]

2015

C

R

U

Malaysia

Comparison with fasciectomy

730

S

1–3

Ext def

24

Waters[31]

2015

CS

R

M

USA

Patients with chronic immunosuppression

201

CORD

S

1

Ext def

24–48

Haerle[37]

2015

CS

P

M

Germany

Open, with no intervention

365

S

1–3

Ext def

24

Stromberg[25]

2016

C

P

U

Sweden

Comparison with aponeurotomy (only MCP)

365

CORD

S

(1–2)

Ext def

24

Murphy[29]

2016

CS

P

U

UK

Cost analysis

690

CORD

S

1.1

(1–2)

Ext def

48

Odinsson[50]

2016

CS

P

U

Norway

Follow-up

365

S

0.8

(1–3)

Ext def

48

Arora[58]

2016

CS

R

U

Austria

Follow-up

365

CORD

S

1–3

Ext def

24

Hirata[54]

2016

CS

P

M

Japan

Open

365

CORD

S

1.2

(1–3)

ROM

24

Verstreken[55]

2016

CS

P

M

Belgium

Multicenter

90

CORD

S

(1–3)

ROM

Malafa[61]

2016

CS

R

U

USA

Follow-up

180

CORD

S

1

(1–3)

Ext def

24

Scherman[26]

2016

C

P

M

Sweden

Comparison with aponeurotomy

365

S

1

Ext def

24–72

Bear[34]

2017

CS

P

M

USA - Australia - Europe

Assessment of new treatments after recurrence

365

CORD

S

1.23

(1–3)

ROM

24

Lauritzon[48]

2017

CS

P

U

Sweden

Follow-up

730

AV

1

Ext def

24–48

Grandizio[49]

2017

CS

R

S

USA

One dose in one joint and the rest of the vial in another cord

30

CORD

AV

1

Ext def

24

Hansen[59]

2017

CS

P

S

Denmark

Comparison of the results between MCF and IFP

365

CORD

S

1.2

(1–3)

Ext def

24

Keller[60]

2017

CS

P

S

Austria

Follow-up

365

S

(1–3)

Ext def

24

Skov[27]

2017

C

P

S

Denmark

Comparison with aponeurotomy (only IFP)

730

S

1

Ext def

24

Table 2

Article

Date

No CCH

Mean age

Male (%)

Previous TRT

Fam His

Hurst[6]

2009

203

62.7

171

79

85

Gilpin[14]

2010

66

63.8

39

28

22

Spanholtz[18]

2011

8

62.5

6

Coleman[57]

2012

12

63.7

11

5

6

Peimer[9]

2013

643

66

542

278

Peimer[65]

2013

463

65.5

342

Vollbach[19]

2013

14

Skirven[32]

2013

21

63

19

12

Marmol[43]

2013

15

64

McMahon[35]

2013

64

N/A

31

Alberton[44]

2013

40

66

36

Martin-Ferrero[45]

2013

35

68.1

35

3

Nydick[24]

2013

29

67

25

Sanjuan[28]

2013

91

65.1

38

Witthaut[15]

2013

587

63.7

498

308

245

Binter[51]

2014

37

66

32

0

Mickelson[40]

2014

43

64.5

35

10

19

Considine[52]

2014

10

66

10

2

Garcia-Olea[64]

2014

148

64

140

24

44

Muppavarapu[20]

2014

73

64

61

24

Sood[30]

2014

16

69.9

16

2

Coleman[46]

2014

60

64

51

26

25

Leclere[33]

2014

33

64.4

28

Povlsen[21]

2014

10

Manning[41]

2014

45

63

33

8

Warwick[38]

2014

144

N/A

119

Verheyden[39]

2014

40

66

38

Perez-Giner[53]

2015

10

65.6

7

Considine[52]

2015

104

61

83

27

56

Zhou[22]

2015

37

65.5

34

Kaplan[42]

2015

714

64

616

376

337

Gaston[17]

2015

164

70

134

23

Atroshi[47]

2015

29

65

13

Tay[23]

2015

237

64

Waters[31]

2015

8

66

Haerle[37]

2015

86

65.1

69

28

Stromberg[25]

2016

69

66

56

0

34

Murphy[29]

2016

20

64.8

20

11

Odinsson[50]

2016

77

69

66

Arora[58]

2016

120

62

Hirata[54]

2016

77

68

70

35

Verstreken[55]

2016

104

64.4

85

85

47

Malafa[61]

2016

36

65.1

33

Scherman[26]

2016

56

36

Bear[34]

2017

52

66.5

50

52

Lauritzon[48]

2017

48

68

38

6

Grandizio[49]

2017

34

65

23

5

Hansen[59]

2017

212

66

195

0

Keller[60]

2017

120

62

107

Skov[27]

2017

29

62

45

The mean age of the patients was 65.2 years (range 61.0–70.0). Male patients corresponded to 85% of the patients in the studies (mean 92.5, range 7–542). Thirty-three series were developed in a single center and 18 studies were multicenter projects. With regard to the temporal evolution, 36 studies were prospective and 15 were retrospective.

We have found a great heterogeneity of studies with many methodological variations. The two clinical trials corresponded to the CORD I[6] and CORD II[14] clinical trials performed in the USA and Australia, respectively, and which compared the effect of CCH over placebo. The nine case-control series compared the effect of CCH with that of surgery through fasciectomy[19] [20] [21] [22] [23] or with needle aponeurotomy.[24] [25] [26] [27] Even though most of the studies were based on clinical series with a time-based follow-up, they ranged from cost studies,[28] [29] implementation of treatment in selective population series,[30] [31] physiotherapy protocols as a comparative group,[32] assistance with ultrasound treatment[33] or the selective reimplementation of treatment to patients who had undergone a previous treatment with CCH.[34] Two studies were limited to assessing the treatment of a single joint of the same finger comparing it with aponeurotomy: Stromberg,[25] on the MCP, and Skov,[27] on the PIP. Skirven[32] performed his study only with PIP results.

As for the study methodology, we have found different models of study design, such as case study series with different programs;[35] [36] open label studies, such as JOINT[15] or ReDUCTo,[37] or studies such as CORD II,[14] that begin as double blind randomized and continue as open label, from which only the data from the first part was collected to include the results of the second part in the CORDLESS study.[9] Warwick[38] performed a series in which results regarding “unusual” cords (natatory, Y-shaped or crow-foot cords) were specified in the same way as Verheyden.[39] Several studies[40] [41] [42] assessed the impact with respect to the extension time following consistent administration of treatment.

The inclusion and exclusion criteria for the studies included were based on the drug's fact sheet. All patients included presented with a minimum initial contracture of 20° in the PIP or MCP joints. Studies that have strictly followed the CORD criteria have limited the maximum degree of flexion also included in the study. The follow-up criteria showed a great deal of heterogeneity. After reading the articles, 26 studies, directly or indirectly, followed the criteria of the CORD clinical trials to demonstrate their results. On the other hand, 5 followed the Tubiana classification,[18] [28] [43] [44] [45] and 1 followed its own criteria.[30] The remaining 19 have not specified their follow-up criteria. The measurement of results has mixed values at varying degrees of measuring: a priori 16 articles used the range of movement (ROM) to evaluate results in whole or in part, while the remaining 35 did not use this method, limiting themselves to assessing extension deficit results in degrees by separate joints, groups of joints (passive extension deficit [PED]), or one article even[44] includes the PIP (total passive extension deficit [TPED]).

The 50 selected clinical trials encompass a total of 4,622 patients (mean 92.70). A total of 7,546 joints were treated with an average of 148.15 joints treated per study (3,925 MCP [10.31 on average] and 2,350 PIP [58.03 on average]). The administered dose was standard in all studies, except for two studies that used double dose on the same hand at a time,[17] [46] and four that used the standard dose plus the amount remaining in the vial in various formats.[39] [47] [48] [49] Twenty-three studies followed an injection protocol in which they allowed one to three infiltrations per joint and patient, four[25] [29] [38] [47] studies used one or two injections, even though they followed the CORD protocols, and presumably if it were needed, they would have used three, and the rest used a protocol of one injection for an infiltration of a joint. In the studies that specified how many vials were used per joint, the mean was 1.23 (range 0.8–1.6; SD: 0.36), with only one study having used less than one vial per joint,[50] and with the studies by Gilpin[14] (1.5) and Sood[30] (1.6) being the ones that used more than one vial per joint. The extension time varied among the studies, ranging from 24 hours (predominantly) up to 7 days in a clinical trial by Mickelson.[40]

The systematic implementation of a physiotherapy protocol after CCH treatment is another controversial point regarding the outcomes. Thirteen articles indicate that patients were systematically referred to an established physiotherapy protocol or that monitoring was performed by specialist physiotherapists,[20] [25] [29] [32] [34] [35] [40] [41] [44] [48] [51] [52] [53] four specify that patients did not receive physical therapy,[6] [22] [49] [54] two have indicated it only if necessary,[46] [55] and the rest did not indicate whether or not a physiotherapy protocol was included in the treatment protocol.

Using night-time splints is recommended for a period of 3 months in the product fact sheet.[56] The bibliography is variable in this regard. Twenty-seven articles refer to the systematic use of this device,[6] [14] [15] [17] [20] [22] [25] [26] [30] [33] [35] [36] [38] [39] [40] [41] [44] [46] [47] [48] [49] [51] [54] [57] [58] [59] [60] and a multicenter study reports that only one of the centers used it without performing a statistical analysis of what this measurement could indicate.[61]


#

Clinical Findings

The studies analyzed indicate a timeline regarding the progression time. In spite of not having a formal indication as to the progression time, the studies mark a before and an after with a timeline of less or more than 1 year. Studies with a course of less than 1 year set out to assess the clinical effectiveness of the treatment or some of its modifications (if the treatment actually works by reducing contracture in DD), and the studies that took 1 year or longer try to assess the effectiveness of the treatment (if the achieved effect is maintained over time). To calculate the clinical outcomes ([Appendix A]), data from studies that lasted less than 1 year ([Table 3]) have been taken into account together with the figures from the intermediate results of the studies that took 1 year or longer ([Table 4]).

Appendix A

Article

Date

Follow-up

Total no

N MCF

N IFP

PEP

PEP MCF

PEP IFP

SEC

SEC MCF

SEC IFP

Corr mean

Corr MCF

Corr IFP

% Total

% MCF

% IFP

Hurst

2009

90

203

133

(65)

70

(34)

130

(64)

102

(76)

28

(40)

73

(35)

31

(23)

42

(60)

36.7

41

29

79

87

64

Gilpin

2010

90

45

20

(44)

25

(55)

20

(44)

13

(65)

7

(28)

25

(55)

7

(35)

18

(72)

25.4

42

32

70.5

79.5

57.6

Spanholtz

2011

14

16

33

33

32

Coleman

2012

270

36

22

(61)

14

(38)

29.7

29

31

79.7

83.1

74.3

Peimer

2013

30

629

398

(63)

231

(36)

284

(45)

222

(55)

62

(26)

125

(19)

62

(15)

63

(27)

38

34

75

81

66

Vollbach

2013

365

16

33

Skriven

2013

30

22

22

(100)

49

88

Marmol

2013

30

31

24

(77)

7

(22)

31

(100)

24

(100)

7

(100)

44

47

98

100

94

McMahon

2013

450

64

46

(71)

18

(28)

41

47

25

Alberton

2013

180

40

32

(80)

8

(20)

29

(72)

60

41

67

Martin-Ferrero

2013

365

35

30

(85)

5

(14)

56

63

Nydick

2013

180

34

14

(41)

5

(14)

30

24

Sanjuan

2013

180

62

31

(50)

31

(50)

42

32

70

88

52

Witthaut

2013

270

879

531

(60)

348

(39)

497

(56)

369

(69)

128

(36)

175

(19)

101

(19)

74

(21)

55

25

72.6

84

55.2

Mickelson

2014

30

45

24

(53)

21

(46)

22

(48)

14

(58)

8

(38)

22

(48)

10

(41)

12

(57)

45

48

46.5

45

48

Considine

2014

45

13

54

30

Garcia-Olea

2014

365

156

58

(37)

16

(10)

38

50

Coleman

2014

60

120

75

(62)

45

(37)

72

(60)

57

(76)

15

(33)

57

32

27

86

66

Leclere

2014

300

52

28

21

Povlsen

2014

10

20

17

Manning

2014

98

50

42

(84)

8

(16)

39

(92)

49

74

51

94

50

Warwick

2014

180

521

43

33

Verheyden

2014

60

38

(N/A*)

30

(N/A*)

27

(71)

13

(43)

33

31

85

76

Perez-Giner

2015

90

13

12

(92)

5

(38)

13

(100)

12

(100)

5

(100)

54

30

Considine

2015

45

33

(50)

33

(50)

39

(59)

25

(75)

14

(42)

13

24

Zhou

2015

84

66

29

(N/A)

45

Kaplan

2015

90

1448

896

(61)

552

(38)

437

(30)

579

(64)

158

(28)

458

(31)

221

(24)

237

(42)

70

74

84

60

Gaston

2015

60

159

55

39

28

Atroshi

2015

21

29

21

(72)

8

(27)

33

19

83

32

Waters

2015

201

13

7

(53)

6

(46)

13

38

43

Haerle

2015

365

86

63

(73)

23

(26)

28

20

Stromberg

2016

365

69

69

(100)

61

(88)

61

(88)

50

50

Murphy

2016

690

45

18

Odinsson

2016

365

109

70

(64)

71

(65)

84

(77)

56

(80)

28

(39)

32

23

Hirata

2016

365

77

47

(61)

30

(38)

66 (85)

44

(93)

22

(73)

3

(3)

0

(0)

3

(10)

40

42

35

91

97

83

Verstreken

2016

90

111

74

(66)

40

(36)

72 (64)

58

(78)

14

(35)

28

(25)

23

(31)

5

(12)

45

86.1

Malafa

2016

180

47

40

(85)

7

(14)

36 (76)

40

41

38

Scherman

2016

365

56

40

(71)

16

(28)

45

55

3

75

Lauritzon

2017

730

50

66

49

17

Grandizio

2017

30

34

32 (94)

20

(N/A)

12

(N/A)

83

45

39

95

80

Hansen

2017

365

235

170

(72)

65

(27)

104 (44)

76

(44)

28

(43)

47

47

47

91

76

Table 3

Article

Date

Follow-up

Total no

No MCF

No IFP

PEP

PEP MCF

PEP IFP

SEC

SEC MCF

SEC IFP

Mean int corr

MCF

IFP

Corr mean

Corr MCF

Corr IFP

% Total

% MCF

% IFP

Hurst[6]

2009

90

203

133 (65)

70 (34)

130 (64)

102 (76)

28 (40)

73 (35)

31 (23)

42 (60)

36.7

41

29

79

87

64

Gilpin[14]

2010

90

45

20 (44)

25 (55)

20 (44)

13 (65)

7 (28)

25 (55)

7 (35)

18 (72)

25.4

42

32

70.5

79.5

57.6

Spanholtz[18]

2011

14

16

33

33

32

Coleman[57]

2012

270

36

22 (61)

14 (38)

29.7

29

31

79.7

83.1

74.3

Peimer[65]

2013

30

629

398 (63)

231 (36)

284 (45)

222 (55)

62 (26)

125 (19)

62 (15)

63 (27)

38

34

75

81

66

Skirven[32]

2013

30

22

22 (100)

49

88

Marmol[43]

2013

30

31

24 (77)

7 (22)

31 (100)

24 (100)

7 (100)

44

47

98

100

94

Alberton[44]

2013

180

40

32 (80)

8 (20)

29 (72)

60

41

67

60

59

67

Nydick[24]

2013

180

34

14 (41)

5 (14)

30

24

Sanjuan[28]

2013

180

62

31 (50)

31 (50)

42

32

70

88

52

Witthaut[15]

2013

270

879

531 (60)

348 (39)

497 (56)

369 (69)

128 (36)

175 (19)

101 (19)

74 (21)

55

25

72.6

84

55.2

Mickelson[40]

2014

30

45

24 (53)

21 (46)

22 (48)

14 (58)

8 (38)

22 (48)

10 (41)

12 (57)

45

48

46.5

45

48

Considine[52]

2014

45

13

54

30

Coleman[46]

2014

60

120

75 (62)

45 (37)

72 (60)

57 (76)

15 (33)

57

32

27

86

66

Manning[41]

2014

98

50

42 (84)

8 (16)

39 (92)

49

48

49

74

51

94

50

Warwick[38]

2014

180

521

49

45

43

33

Verheyden[39]

2014

60

38 (N/A)

30 (N/A)

27 (71)

13 (43)

38

36

33

31

85

76

Perez-Giner[53]

2015

90

13

12 (92)

5 (38)

13 (100)

12 (100)

5 (100)

54

30

Considine[52]

2015

45

33 (50)

33 (50)

39 (59)

25 (75)

14 (42)

13

24

Zhou[22]

2015

84

66

29 (N/A)

45

Kaplan[42]

2015

90

1448

896 (61)

552 (38)

437 (30)

579 (64)

158 (28)

458 (31)

221 (24)

237 (42)

27

70

74

84

60

Gaston[17]

2015

60

159

55

39

28

Atroshi[47]

2015

21

29

21 (72)

8 (27)

33

19

83

32

Waters[31]

2015

201

13

7 (53)

6 (46)

13

38

43

Verstreken[55]

2016

90

111

74 (66)

40 (36)

72 (64)

58 (78)

14 (35)

28 (25)

23 (31)

5 (12)

45

86.1

Malafa[61]

2016

180

47

40 (85)

7 (14)

36 (76)

40

41

38

Grandizio[49]

2017

30

34

32 (94)

20 (N/A)

12 (N/A)

83

45

39

95

80

Table 4

Article

Date

Follow-up

Total No

No MCF

No IFP

PEP

PEP MCF

PEP IFP

SEP

SEP MCF

SEP IFP

Mean int corr

MCF

IFP

Corr mean

Corr MCF

Corr IFP

% Total

% MCF

% IFP

Peimer[9]

2013

1095

1080

648

(60)

432

(40)

623

(57)

451

(69)

172

(39)

301

(27)

152

(23)

149

(34)

21

28

15

Vollbach[19]

2013

365

16

33

31

McMahon[35]

2013

450

64

46

(71)

18

(28)

41

47

25

33

42

10

Martin-Ferrero[45]

2013

365

35

30

(85)

5

(14)

56

63

60

58

Binter[51]

2014

365

40

14

(35)

8

(20)

21

(52)

Garcia-Olea[64]

2014

365

156

58

(37)

16

(10)

38

50

35

23

Muppavarapu[20]

2014

426

100

56

(56)

44

(44)

29

(29)

26

(46)

3

(6)

28

21

37

Sood[30]

2014

369

27

18

(66)

9

(33)

27

(100)

18

(100)

9

(100)

36

29

83.3

55.6

Leclere[33]

2014

300

52

33

31

28

21

Tay[23]

2015

730

298

99

(33)

56

(18)

146

(48)

95

(95)

51

(91)

8

(2)

3

(3)

5

(8)

45

56

96

97

91

Haerle[37]

2015

365

86

63

(73)

23

(26)

28

20

31

8

Stromberg[25]

2016

365

69

69

(100)

61

(88)

61

(88)

50

50

47

47

Murphy[29]

2016

690

45

18

35

6

Odinsson[50]

2016

365

109

70

(64)

71

(65)

84

(77)

56

(80)

28

(39)

32

23

28

16

Arora[58]

2016

365

120

85

(70)

31

(25)

28

30

Hirata[54]

2016

365

77

47

(61)

30

(38)

66

(85)

44

(93)

22

(73)

3

(3)

0

(0)

3

(10)

40

42

35

91

97

83

Scherman[26]

2016

365

56

40

(71)

16

(28)

48

55

3

45

50

-2

75

Bear[34]

2017

365

51

31

(60)

20

(39)

18

(35)

11

(35)

7

(35)

32

33

31

83

69

Lauritzon[48]

2017

730

50

66

49

17

59

45

14

Hansen[59]

2017

365

235

170

(72)

65

(27)

104

(44)

76

(44)

28

(43)

47

47

47

47

39

91

76

Keller[60]

2017

365

120

85

(70)

28

18

Skov[27]

2017

730

29

29

(100)

8

(27)

40

8

Studies that Cover Less than 1 Year of Progress

The mean follow-up time was 104 (SD: 75) days. The analysis included a total of 4,666 joints (average of 194.41 joints per study), (2,467 MCPs (average of 129.84) and 1,516 PIPs (average of 75.80)). The primary end point objective has been achieved in 48.9% of the treated joints (69.77% of the MCPs and 30.14% of the PIPs). The mean correction for all joints treated was 45.5 (SD: 19.18) degrees; 40.8 degrees for MCPs (SD: 10.12) and 35.6 for PIPs (SD: 13.23). The proportional correction of the joints was at 72.9% (SD: 14.43) overall, (83.9% for the MCP [SD: 12.58] and 64.2 for the PIP [SD: 16.35]).


#

Studies that Cover More than 1 Year of Progress

The mean follow-up time was 467 days (SD: 196.36). The analysis included a total of 2,870 joints (average of 138.66 joints per study), (1,459 MCPs [average of 97.26] and 842 PIPs [average of 56.13]). The primary end point objective has been achieved in 57.5% of the treated joints (68.9% of the MCPs and 43.3% of the PIPs). The mean correction for all joints treated was 37.6 degrees (SD: 10.93), (37.3 degrees for the MCP [SD: 9.98] and 23.7 for the PIP [SD: 16.33]. The proportional correction of joints was at 87.3% (SD: 10.96) overall, (90.3% for the MCP [SD: 6.94] and 75% for the PIP [SD: 13.54]).


#

Recurrences

Apart from the CORDLESS studies,[9] [10] few studies have collected the incidence of recurrences in their results.[19] [30] [34] [35] [37] [48] [58] The rate of recurrence during the follow-up period ranged from 1.7% in 1 year[37] to 59% in the series by Sood[30] within the same period, but with its own criteria of recurrence. Upon eliminating these two values, this rate ranged between 7–28%. All the articles that mentioned the rate of recurrence had a follow-up between 6 and 18 months. We should keep in mind that the series by Bear[34] had a 28% recurrence rate in patients previously treated with CCH after 1 year of follow-up.

The CORDLESS studies were designed as observational clinical studies for assessing the rate of recurrence over time. These follow-ups were published at 39 and 5 years.[10] The recurrence rate was calculated based on the secondary end point (at least 50% of correction since the initial contracture), demonstrating a recurrence rate of 38% (28% for the MCP and 58% for the PIP) at 3 years, and 48% (39% for the MCP and 65% for the PIP) at 5 years. Peimer indicates that 75% of the recurrences occurred after the first 3 years of treatment[10] and are significantly minor regarding subsequent progression.


#

Treatment Failures

A concept not adequately clarified in the literature regarding CCH treatment is treatment failures, that is, the injection of CCH into a Dupuytren cord with no effect. Peimer,[10] at the 5-year follow-up of the CORDLESS study, indicates that the rate of patients treated ineffectively in the CORD and JOINT studies was 18% (9% of MCPs and 32% of PIPs), whereas among the patients who completed the follow-up for the CORDLESS study, the rate was 14% (7% of MCPs and 26% of PIPs). Only two more authors have covered this concept: Keller[60] indicates a 3% rate of treatment failures, and Peimer (ref), in another study that does not include patients from the CORD studies, reports a rate of 16% (8 patients).


#
#
#

Discussion

Collagenase clostridium histolyticum is currently a therapeutic alternative for DD both in Europe[8] and the USA.[7] However, the heterogeneity of publications has been more the norm rather than the exception in studies related to CCH. Even though virtually all studies have maintained a standard minimum contracture with which to begin the treatment (20 degrees), some have slightly increased this to 30 degrees. Other data, such as the loss of extension limit (initially set at 100 degrees for the finger), the variation in severity criteria with regard to adoption of own criteria rather than any other DD classification, show the heterogeneity in publications, as is common in publications related to Dupuytren surgery.[13]

The measurement of results is another example: the CCH treatment is performed at a joint level in an isolated manner; with the rupture of the cord, a compensatory correction of the adjacent joint of the same finger can be produced. Measuring both joints together after the treatment of only one of them constitutes an error that causes bias in clinical studies, as the correction of a joint can be masked by the affectation or the retraction of the adjacent finger.[62] Likewise, the inclusion of untreated joints in the results, when they present baseline contracture, worsens the final result if a joint finger assessment (MCP + PIP) is done. In fact, to mitigate these shortcomings, in CORD studies the results are presented in two modes: on the one hand in terms of the correction of the treated joint, and on the other in the form of “range of movement” (ROM).[6] [14] [54] Different forms of measurement (ROM, passive extension deficit [PED], isolated joints) are adopted in the studies, which complicate the comparability of results between the studies ([Table 1]).

Clinical results indicate that full extension of the fingers has been achieved in ∼ 50% of the cases maintained for a year, and the mean reduction in contracture for all patients is at around 75%, indicating a remarkable effectiveness of the treatment. The results of our study may seem contradictory with a higher reduction in digital contracture in studies lasting over 1 year than in studies covering less than 1 year of progress, but the heterogeneity of published papers and the lack of homogeneity between the protocols used[63] explain this variation, which does not invalidate the final result. The fact that a greater number of patients have reached the primary end point in studies lasting more than 1 year is due, in part, to the fact that many of these studies have been funded by or related to the company that markets the drug, which is why the patient selection criteria, the surgeon's experience or the protocol for patient maintenance, minimizing the losses of individuals with a positive result, may be regarded as biased.[9] [34] [35] [37] [48] [54] [60]

With regard to joint treatment, the demonstrated results concur with many of the series published. The outcome for the MCP joint is better than the outcome for the PIP, and the recurrences of the latter are also much more frequent. The data presented does not allow for a study of the severity of the affected joint as some authors do[9] [14] [46] [57]; the analyzed data indicate that a more severe initial contracture signifies a worse outcome for the treatment, more specifically for the PIP. With respect to the issue of recurrence, the lack of studies in the medium and long term prevents an objective assessment; however, the CORDLESS studies show a clear tendency toward recurrence with an overall rate of 48% at 5 years.[10] Garcia-Olea[64] quantifies the deterioration of the patients in their series at 1.5 degrees per month.

Physiotherapy protocols have not been included in clinical trials; thus, the improvement or sustainability of treatment with them cannot be assessed. Although the use of night splints is indicated, many studies have avoided this adjunctive measure, possibly due to poor compliance of the patients,[41] and the discomfort it causes over an extended period of time. These two measures, without a doubt, are the most variable in the studies analyzed. One of the first comparative clinical studies[32] actually assessed the use of these measures for treatment with CCH and concluded that their apparent benefit is in the short term.

Among the main problems that are not taken into account in the analyzed series is the rate of poor and “non-effective” outcomes. The same analysis is not performed on these patients as it is performed on those with good outcomes. There is currently no explanation as to why some patients do not respond to the treatment.

Among the limitations of our study, the main one is the possibility of biases in terms of data collection. Injection on a joint and the measurement of the full result on the affected finger is in itself a bias that some of the cited papers comment on, and it has made it difficult to compare this technique with the surgical procedures (partial fasciectomy), where traditionally the whole radius is treated. The results are usually expressed as a difference in degrees between the initial and final results with a range in each value. While obtaining the difference between both results for the assessment of the correction achieved is very simple, it is practically impossible to evaluate the standard deviation of the sample. This prevents large-scale statistical studies from being conducted, and thus constitutes the main limitation of our study. Unfortunately, these limitations are insurmountable and evident in the analysis of bibliography.[13]

In conclusion, we can say that there is a lack of uniformity in the approach of the studies and in their outcomes with regard to the assessment of DD treatment with CCH. Despite this issue, the results indicate a satisfactory response to treatment in a large number of patients maintained in the short and medium term, with a recurrence rate of over 50%, occurring mainly within the first 3 years of follow-up. The exact recurrence rate is uncertain given the available data.


#
#

Conflicts of Interest

Authors declare that there were no conflict of interests for the writing of the present work.


Address for correspondence

Rafael Sanjuan Cerveró, MS
Department of Orthopedics and Traumatology Surgery
Hospital de Dénia, Partida de Beniadla s/n, 03700, Dénia, Alicante
Spain   


Zoom Image
Fig. 1 Flowchart of the selection of articles analyzed.
Zoom Image
Fig. 1 Diagrama de flujo de la selección de artículos analizados.