Low molecular weight heparin for prevention of microvascular occlusion in digital replantation

  • Review
  • Intervention

Authors

  • Yi-Chieh Chen,

    1. Chang Gung University, College of Medicine, Taoyuan, Taiwan
    2. Chang Gung Memorial Hospital, Department of Plastic and Reconstructive Surgery, Taoyuan, Taiwan
    Search for more papers by this author
  • Ching-Chi Chi,

    1. Chang Gung University, College of Medicine, Taoyuan, Taiwan
    2. Chang Gung Memorial Hospital, Department of Dermatology and Centre for Evidence-Based Medicine, Chiayi, Taiwan
    Search for more papers by this author
  • Fuan Chiang Chan,

    Corresponding author
    1. Temple Street Children's University Hospital, Department of Plastic and Reconstructive Surgery, Dublin, Ireland
    • Fuan Chiang Chan, Department of Plastic and Reconstructive Surgery, Temple Street Children's University Hospital, Temple Street, Dublin, Ireland. fchan910@gmail.com. fchan@eircom.net.

    Search for more papers by this author
  • Yu-Wen Wen

    1. Chang Gung University, Clinical Informatics and Medical Statistics Research Centre, Taoyuan, Taiwan
    Search for more papers by this author

Abstract

Background

The success of digital replantation is highly dependent on the patency of the repaired vessels after microvascular anastomosis. Antithrombotic agents are frequently used for preventing vascular occlusion. Low molecular weight heparin (LMWH) has been reported to be as effective as unfractionated heparin (UFH) in peripheral vascular surgery, but with fewer adverse effects. Its benefit in microvascular surgery such as digital replantation is unclear.

Objectives

To assess whether subcutaneous LMWH treatment improves the salvage rate of the digits in patients with digital replantation after traumatic amputation.

Search methods

The Cochrane Peripheral Vascular Diseases Group Trials Search Co-ordinator (TSC) searched the Specialised Register (October 2012), CENTRAL (2012, Issue 10) and trials databases. In addition, the authors searched PubMed, CNKI (China National Knowledge Infrastructure) and CEPS (Chinese Electronic Periodical Services), and sought additional trials from reference lists of relevant publications.

Selection criteria

We selected randomised or quasi-randomised controlled trials of LMWH in patients who received digital replantation.

Data collection and analysis

Two review authors independently extracted data and assessed the risk of bias of the included trials. Disagreements were resolved by discussion.

Main results

Two randomised trials involving 114 patients with at least 122 replanted digits met the inclusion criteria and were included. Both trials compared the efficacy and safety of LMWH with UFH. We found no trials comparing LMWH with placebo or other anticoagulants. The data from the two included studies were insufficient for meta-analysis. The overall success rate of replantation did not differ between the LMWH and UFH groups, 92.3% versus 89.2% in one trial (risk ratio (RR) 1.03; 95% confidence interval (CI) 0.87 to 1.22) and 94.3% versus 94.15% in the other trial (RR 1.00; 95% CI 0.89 to 1.13). The incidence of both postoperative arterial and venous insufficiency were reported in one trial and did not significantly differ between the LMWH and UFH groups (RR 1.08; 95% CI 0.16 to 7.10 and RR 0.81; 95% CI 0.20 to 3.27, respectively). Direct and indirect causes of microvascular insufficiency were not reported in the trials. Different methods were used to monitor the adverse effects related to anticoagulation in the two trials. Bleeding tendency was monitored for the LMWH and UFH groups in one trial and was reported by the incidence of wound haemorrhage (11.5% versus 17.9%; RR 0.65; 95% CI 0.17 to 2.44), ecchymoses (3.8% versus 10.7%; RR 0.36; 95% CI 0.04 to 3.24), haematuria (3.8% versus 7.1%; RR 0.54; 95% CI 0.05 to 5.59), nasal bleeding (0% versus 7.1%; RR 0.21; 95% CI 0.01 to 4.28), gingival bleeding (0% versus 10.7%; RR 0.15, 95% CI 0.01 to 2.83) and faecal occult blood (0% versus 3.6%; RR 0.36; 95% CI 0.02 to 8.42). The bleeding tendency was increased in the UFH group but this was not statistically significant. This trial also monitored coagulability changes using parameters such as antithrombin activity, factor Xa activity, bleeding time, clotting time and activated partial thromboplastin time (aPTT). No comparison was made between the LMWH and UFH groups but all data consistently showed that coagulability was reduced more in the UFH group than in the LMWH group. The other trial reported a postoperative decrease in platelet count in the UFH group (preoperative 278.4 ± 18.7 x 109/L, postoperative 194.3 ± 26.5 x 109/L; P < 0.05) but not in the LMWH group (preoperative 260.8 ± 32.5 x 109/L, postoperative 252.4 ± 29.1 x 109/L; P > 0.05).

Authors' conclusions

Current limited evidence based on two small-scaled low-to-medium quality randomised trials found no differences in the success rate of replantation between LMWH and UFH, but a lower risk of postoperative bleeding and hypocoagulability after the use of LMWH. Further well-designed and adequately powered clinical trials are warranted.

Plain language summary

Low molecular weight heparin for preventing vascular occlusion in replanted fingers or toes after amputation injuries

Microvascular surgery refers to any surgery involving small sized blood vessels that is performed under the operating microscope, allowing the repair of arteries and veins of the digits. These are typically 1 mm to 2 mm in diameter. Replantation is the reattachment of a completely detached body part, with fingers and thumbs being the most commonly replanted body parts. This is often referred to as digital replantation. In principle, digital replantation involves not only restoring the blood flow through the arteries and veins but also restoring the bony skeleton of the toes, fingers or thumbs, along with repairing the tendons and nerves as indicated. Occlusion of one or more of the repaired vessels due to the formation of a clot (thrombus) within the blood vessel results in failure of the replantation. Anticoagulant medications are used to reduce clotting, and they could potentially prevent such a complication. Anticoagulants such as unfractionated heparin (UFH) have therefore been used to prevent clot formation after digital replantation. It is unclear if low molecular weight heparin (LMWH) has similar benefit. This systematic review identified only two randomised controlled trials comparing UFH with LMWH, with a total of 114 patients. No studies were identified that compared LMWH with placebo, no treatment or other anticoagulants. The limited data from the two trials showed no difference between LMWH and UFH in the success rate for digital replantation but there were less frequent anticoagulation-related adverse events (such as bleeding) with LMWH. The available evidence is insufficient to make a firm conclusion.

Background

Description of the condition

Digital replantation is an established microsurgical procedure in traumatic hand surgery, with reported success rates as high as 90% (Levin 2008). The success is highly dependent on maintaining the patency of the repaired blood vessels after satisfactory microvascular anastomosis. Microvascular occlusion can result from either venous or arterial thrombosis or, less frequently, a combination of both. Arterial thrombi, which usually present in areas of high or disturbed blood flow or at sites of endothelial damage, are formed primarily of platelet aggregates bound together by thin fibrin strands; thus both platelet activation and the coagulation cascade are important in its thrombogenesis. Venous thrombi, on the other hand, present in areas of stasis and are formed primarily by red blood cells and fibrin, and less so by platelets; thus the coagulation cascade plays a much more prominent role than platelet activation in their etiology (Conrad 2001). Studies on replantation of the digits and free-tissue transfer have indicated that the highest risk of critical thrombosis following microvascular anastomosis is in the first three days following surgery. The risk is then reduced but still exists up to two weeks after surgery (Betancourt 1998; Kroll 1996). Venous thrombosis is reported to occur more often than arterial thrombosis. Nevertheless, 90% of arterial thrombi occur within 24 hours post-replantation while the majority of venous thromboses tend to occur after the first 24 hours (Askari 2006; Kroll 1996; Levin 2008).

The use of prophylactic antithrombotic agents is the most commonly reported strategy for avoiding vascular thrombosis after vascular repair. A variety of anticoagulation protocols are used by microsurgeons. Current protocols include the use of antithrombotic agents such as aspirin, intravenous heparin, low molecular weight heparins (LMWHs) or intravenous dextran, as well as local heparin delivery through direct injection or continuous drip on an open incision (Askari 2006; Barnett 1989; Buckley 2011; Iglesias 1999). However, the optimum dosage and duration of administration of these agents remains complicated by the wide variability of antithrombotic prophylaxis protocols practiced among different surgical units.

Description of the intervention

Unfractionated heparin (UFH) is a glycosaminoglycan polymer of varying lengths and is the most widely used anticoagulant agent for preventing both arterial and venous thrombosis (Stockmans 1997). Heparin maintains the patency of microvascular anastomoses because it inhibits the synthesis of thrombin. This in turn leads to an improved survival rate of digital replantation. However, heparin prophylaxis is limited by an increased risk of haemorrhage from the surgical site, formation of haematoma, heparin-induced thrombocytopenia (Chong 1989) and requirement for transfusion (Isaacs 1977). Low-dose regimens are attractive from the standpoint of bleeding complications (Kroll 1995), but they have been reported as less effective compared to full heparinising doses (Hendel 1984). On the contrary, LMWH is a group of antithrombotic agents with different individual properties. LMWH is derived from UFH and has been reported to be as efficacious as heparin in preventing thrombosis. It has enhanced bioavailability as it binds loosely to protein and interacts less with platelets. In comparison to UFH it has fewer adverse effects. The LMWHs have a long half-life with a very predictable dose-response relationship, which minimises the need for monitoring. Furthermore, LMWH is associated with reduced risk of bleeding and a lower risk of heparin-induced osteoporosis. In addition, it has been shown that short-term use of LMWH is less frequently associated with undesirable heparin-induced thrombocytopenia (HIT) compared with UFH (Franklin 2003; Warkentin 1995).

How the intervention might work

When the vessel wall is injured, collagen and tissue factor become exposed to the flowing blood thereby initiating a proteolytic cascade that leads to the formation of a thrombus (Furie 2008). Exposed tissue factor triggers the formation of thrombin. Unlike the formation of thrombi in large sized vessels, thrombus formation in small sized vessels, such as the digital artery, may completely occlude the blood flow to the tissues supplied by the artery. Similarly, thrombus formation in the digital vein impedes proper venous outflow, which will eventually lead to arterial thrombosis. Both UFH and LMWH exert their anticoagulant activity by activating antithrombin (also known as antithrombin III), which in turn inhibits thrombin and activated factor Xa (factor Xa) to prevent both arterial and venous thrombosis (Hirsh 2001; Wolf 1994). Although systemic anticoagulant therapy does not improve the patency rate when sharply divided vessels are repaired (Elcock 1972; Ketchum 1978), a laboratory study showed a dramatic improvement in the patency rate following repair of traumatized vessels using systemic anticoagulant as an adjunct (Cooley 1985). In a retrospective study of replantation failure, 20% occurred within four hours of discontinuing systemic heparin (Hendel 1984). Although re-endothelialisation begins immediately after vascular repair, some form of anticoagulant therapy should be continued for at least three to five days, that is until the endothelium regenerates and covers the anastomotic site (Chow 1983; Morecraft 1985; Servant 1976). The LMWHs are produced by either chemical or enzymatic depolymerization of UFH resulting in the same inhibitory effect on active factor X but with relatively less anti-IIa (thrombin) activity and thus less effect on the clotting time than UFH (Weitz 1997).

Why it is important to do this review

There are conflicting reports on both the efficacy and adverse effects of anticoagulants in preventing microvascular thrombosis (Niibayashi 2000; Veravuthipakorn 2004; Vretos 1995) with little objective evidence to demonstrate their beneficial effects in digital replantation. In addition, many surgeons favour their own particular protocols that are derived by trial and error for perioperative anticoagulation. No consensus or standard exists on the use of LMWH as an antithrombotic agent. As such, there is a strong need to review the efficacy of LMWH in digital replantation.

Objectives

To assess whether subcutaneous low molecular weight heparin (LMWH) treatment improves the salvage rate of the digits and survival in patients with digital replantation after traumatic amputation.

Methods

Criteria for considering studies for this review

Types of studies

We limited our analysis to randomised controlled trials (RCTs), with or without blinding. We also included studies which used alternative methods of randomisation such as alternate days of the week, odd or even dates of birth, or hospital number (quasi-randomised studies). We excluded studies that used historical controls.

Types of participants

We included all patients who suffered from traumatic digital amputation and received salvage microvascular replantation.

Types of interventions

We included studies in which patients with traumatic digital amputation were randomised to receive LMWH versus any other treatment (such as another anticoagulant, placebo or no intervention) after digital replantation in order to prevent microvascular occlusion. We excluded trials that used LMWH for therapeutic purposes in established microvascular occlusion of replanted digits.

Types of outcome measures

Primary outcomes
  • The success rate of replantation, that is, the patency rate of microvascular anastomosis.

  • The incidence of compromised microcirculation requiring surgical re-exploration or re-anastomosis, or both.

Secondary outcomes
  • Direct cause of microvascular insufficiency (arterial occlusion, venous occlusion, or both).

  • Indirect cause of microvascular insufficiency (shock, infection, peripheral vascular pathology, hypercoagulopathy, etc.).

  • Complications and side effects related to the interventions (haemorrhage, thrombocytopenia, etc.).

  • Coagulation abnormalities.

Search methods for identification of studies

The searches were not limited by language or publication status.

Electronic searches

The Cochrane Peripheral Vascular Diseases Group Trials Search Co-ordinator (TSC) searched the Specialised Register (last searched October 2012) and the Cochrane Central Register of Controlled Trials (CENTRAL) (2012, Issue 10), part of The Cochrane Library at www.thecochranelibrary.com. See Appendix 1 for details of the search strategy used to search CENTRAL. The Specialised Register is maintained by the TSC and is constructed from weekly electronic searches of MEDLINE, EMBASE, CINAHL, AMED, and through handsearching relevant journals. The full list of the databases, journals and conference proceedings which have been searched, as well as the search strategies used, are described in the Specialised Register section of the Cochrane Peripheral Vascular Diseases Group module in The Cochrane Library (www.thecochranelibrary.com).

The following trial registers were searched by the TSC for ongoing and unpublished trials (November 2012) using the terms (finger or digit) and heparin:

The review authors also searched PubMed using the search strategy which is detailed in Appendix 2, as well as CNKI (China National Knowledge Infrastructure) at http://cnki50.csis.com.tw/kns50/index.aspx and CEPS (Chinese Electronic Periodical Services) at http://www.airitilibrary.com/ using Chinese synonyms modified from the PubMed search strategy (Appendix 3; Appendix 4). PubMed, CNKI and CEPS were searched on 19 January 2013.

Searching other resources

We examined the reference lists of relevant review articles and all included trials to identify further studies.

Data collection and analysis

We considered all the trials identified with our search strategy for inclusion in this review.

Selection of studies

Two review authors (YCC, FCC) independently assessed all titles and abstracts of potentially eligible trials. We obtained the potentially relevant papers and assessed their full texts for inclusion eligibility. We resolved disagreements by discussion.

Data extraction and management

Two review authors (YCC, FCC) independently extracted and checked the data for accuracy. If relevant data could not be extracted, we tried to contact the primary authors of any articles and requested and included the additional unpublished data, when available. We resolved any discrepancies by consensus.

We used a standard data extraction form to capture the following information:

1. characteristics of the study, including design, method of randomisation, allocation concealment, withdrawals or dropouts, and funding source;
2. study participants, including mechanism of amputation and level of amputation;
3. intervention (dosage and route of LMWH administration);
4. comparison intervention (placebo or other anticoagulant medication);
5. outcome measures, including replantation success rate (patency rate of microvascular anastomosis), incidence of microvascular insufficiency demanding surgical re-exploration, cause of microvascular occlusion, and complications.

Assessment of risk of bias in included studies

Two review authors (YCC, FCC) separately assessed the quality of the included studies using the following key domains: sequence generation, allocation concealment, blinding, incomplete outcome data, selective reporting, and other sources of bias. For each of these domains we assigned a judgement of 'low risk' of bias, 'high risk' of bias, or 'unclear risk' of bias using guidance from the Cochrane Collaboration risk of bias tool (Higgins 2011). We resolved any disagreement by discussion. We presented this information in the Characteristics of included studies table.

Measures of treatment effect

We expressed the data as risk ratio (RR), risk difference (RD), number needed to treat (NNT) and mean difference (MD), where appropriate. The 95% confidence intervals (CI) were used for these estimates of treatment effects.

Unit of analysis issues

We expected to find only studies that used randomisation by individual patients. If studies which randomised by digits had been found, we would apply separate analyses.

Dealing with missing data

We tried to contact the original authors and request the essential information whenever possible. If we were unable to collect the missing data, we would include this fact in the risk analysis.

Assessment of heterogeneity

We planned to examine between-study heterogeneity by inspecting the forest plots and quantifying the impact of heterogeneity using the I2 statistic. An I2 value greater than 50% indicates the possible presence of heterogeneity. If statistical heterogeneity had existed in relationship to the study quality, participants, intervention regimens or outcome measurements we planned to apply a subgroup analysis to handle the heterogeneity; if the heterogeneity was partly due to sample selection we planned to use a random-effects model.

Assessment of reporting biases

When at least 10 relevant trials for a primary outcome were available, we planned to use funnel plots to investigate the presence of publication bias in the included studies, or if any systematic differences existed between small and large studies.

Data synthesis

If appropriate, we planned to perform meta-analyses using the Review Manager software (RevMan 5) supplied by The Cochrane Collaboration. For estimates of the RR and RD, we planned to use the Mantel-Haenszel method. For continuous variables, we planned to use the inverse variance method. We planned to use a fixed-effect model for the meta-analyses except in cases where heterogeneity was identified. In those cases we planned to use a random-effects model as described in the Assessment of heterogeneity section.

Subgroup analysis and investigation of heterogeneity

We planned to perform subgroup analyses according to the mechanism of digital amputation (for example guillotine, crush, avulsion or degloving) and the level of digital amputation when data were available. We also planned to perform a comparison of effects of different heparin doses if sufficient numbers of studies, or trials using different dosage regimens, were identified.

Sensitivity analysis

We planned to perform sensitivity analysis to explore the influence of the quality of studies on the treatment effect size. We planned to use an intention-to-treat (ITT) approach. We planned to analyse only the data available and consider the dropout rate as a marker of trial quality.

Results

Description of studies

See Characteristics of included studies and Characteristics of excluded studies.

Results of the search

See Figure 1.

Figure 1.

Study flow diagram.

Included studies

Two RCTs published in Chinese met our inclusion criteria and were included in this review. Both RCTs (Chen 2001; Li 2012) compared LMWH with UFH. No RCTs comparing LMWH with placebo, no treatment or other anticoagulants were identified.

The first trial (Chen 2001) enrolled 54 participants, including 39 single digit amputations, seven multiple digit amputations, three palm amputations, two forearm amputations and three toe amputations. The inclusion and exclusion criteria were not reported. Twenty-six participants were randomised to the LMWH group (Livaracine 5000 IU subcutaneous injection (SC) 30 min before surgery, followed by 2500 to 5000 IU SC every 12 hours (q12h) for 7 days); the other 28 were randomised to the UFH group (UFH 2500 IU intravenous injection (IV) 0.5 to 1 hour before surgery, followed by 1250 IU q12h for 7 to 10 days). The UFH group also received 500 ml low molecular weight dextran-40 twice a day. No participants with a previous haemorrhagic disorder or preoperative coagulation test that was abnormal were found in either group. The details of the surgical interventions were not reported.

The second trial (Li 2012) enrolled 60 participants with 69 either complete or incomplete amputated digits. The levels of amputation were located between the metacarpophalangeal joint and distal phalangeal base of the digit, and the digital arteries and veins were repaired by interrupted sutures using 9-0 to 12-0 nylon under surgical microscope. In the LMWH group (30 participants with 35 replanted digits), the repair of one artery and two veins was performed on 26 digits of 23 participants; the repair of one artery and one vein was performed on three participants and three digits; and the repair of two arteries and three veins was performed on six digits of four participants. In the UFH group (30 participants with 34 replanted digits), repair of one artery and two veins was performed on 25 digits of 22 participants; repair of one artery and one vein was performed on two participants and two digits; and repair of two arteries and three veins was performed on seven digits of six participants. The participants in the LMWH group received LMWH (exact type unknown) 0.4 ml SC q12h for 7 to 10 days, and participants in the UFH group received UFH 10000 IU SC q12h for 7 to 10 days.

See the Characteristics of included studies tables for further details.

Excluded studies

Twenty reports of studies were excluded after examination of the full texts. See the Characteristics of excluded studies for further details.

Risk of bias in included studies

We considered the Chen 2001 trial as having a high risk of bias and the Li 2012 trial a moderate risk of bias based on the various assessed parameters of methodological quality. Details of the methodological quality are reported in the table Characteristics of included studies. A risk of bias summary is presented in Figure 2 with each methodological quality item presented as a percentage across all included trials; Figure 3 shows the separate judgements for each included trial.

Figure 2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

The method of allocation concealment was not mentioned in either included trial, and the risk of bias for allocation concealment was therefore assessed as unclear.

Blinding

One trial had a high risk of bias for blinding because LMWH was administered subcutaneously but UFH was administered intravenously (Chen 2001), while the other trial had a low risk of bias for blinding because both the participants and personnel were blinded (Li 2012).

Incomplete outcome data

Both included trials were classified as at unclear risk of bias for incomplete outcome data due to a lack of descriptions about dropouts or withdrawals.

Selective reporting

One trial was classified as unclear for risk of selective reporting bias (Li 2012) because the trialists did not report the incidence of compromised microcirculation requiring surgical re-exploration or re-anastomosis. The direct or indirect cause of microvascular insufficiency was also not provided. The other trial (Chen 2001) did not provide detailed statistics (for example SD) regarding coagulability and platelet counts and was classified as having a high risk of reporting bias.

Other potential sources of bias

One trial had a high risk of bias for other sources (Li 2012) and the other trial had an unclear risk of bias for other sources (Chen 2001). The Li 2012 trial did not report which LMWH product was used, the dosage of LMWH, the mechanism of injury, and the severity of soft tissue damage. The participants had a varying number of repaired vessels. All of these factors may have led to a biased effect estimate. The Chen 2001 trial included amputations of the palm and forearm but the number was small (n = 5). Low molecular weight dextran was used only in the UFH group, which might have biased the results. The details are reported in the 'Risk of bias' tables under Characteristics of included studies.

Effects of interventions

Low molecular weight heparin (LMWH) versus unfractionated heparin (UFH)

Both included trials (Chen 2001; Li 2012) compared LMWH and UFH and reported the efficacy and adverse effects. However, the available data were inadequate for meta-analysis because the trialists used different units of analysis to report the success rate of replantation. In addition, there was a lack of relevant data for the other outcomes planned in this review. For the same reason, subgroup and sensitivity analyses were not performed.

Primary outcomes
Success rate of replantation

Both RCTs reported data on this outcome. The Chen 2001 trial reported the outcome by considering individual participants: the success rate of replantation was 92.3% in the LMWH group and 89.2% in the UFH group (RR 1.03; 95% CI 0.87 to 1.22). The Li 2012 trial reported this outcome by considering the digits: the success rate of replantation was 94.3% in the LMWH group and 94.15% in the UFH group (RR 1.00; 95% CI 0.89 to 1.13). In both RCTs the success rate of replantation did not significantly differ between the LMWH and UFH groups (see Analysis 1.1). However, neither of the trialists provided the power of the tests.

Incidence of compromised microcirculation

The Chen 2001 trial reported the incidence of postoperative arterial and venous insufficiency separately (see Analysis 1.2), but the authors did not clarify their definitions for these complications. Arterial insufficiency occurred in two out of 26 participants (7.7%) in the LMWH group and two out of 28 participants (7.1%) in the UFH group (RR 1.08; 95% CI 0.16 to 7.10). Compromised venous drainage occurred in three out of 26 participants (11.5%) in the LMWH group and four out of 28 participants (14.3%) in the UFH group (RR 0.81; 95% CI 0.20 to 3.27). The incidence of arterial and venous insufficiency did not significantly differ between the LMWH and UFH groups. A power analysis was not done. The trialists did not report any cases receiving surgical re-exploration or re-anastomosis for salvage of the condition. The Li 2012 trial did not report data relevant to this outcome.

Secondary outcomes
Direct and indirect cause of microvascular insufficiency

Not mentioned in either RCT.

Complications and side effects related to the interventions
Haemorrhage

The Chen 2001 trial monitored anticoagulation-related bleeding tendency and reported the respective number of participants with various types of haemorrhage (Analysis 1.3) instead of the number of participants with any haemorrhage. Wound haemorrhage occurred in three participants (11.5%) in the LMWH group and five participants (17.9%) in the UFH group (RR 0.65; 95% CI 0.17 to 2.44). Ecchymoses occurred in one participant (3.8%) in the LMWH group and three participants (10.7%) in the UFH group (RR 0.36; 95% CI 0.04 to 3.24). Haematuria appeared in one participant (3.8%) of the LMWH group and two participants (7.1%) of the UFH group (RR 0.54; 95% CI 0.05 to 5.59). Additionally, nasal bleeding, gingival bleeding, and faecal occult blood were present in two (7.1%), three (10.7%), and one participant (3.6%) of the UFH group, respectively; but none of the LMWH group. Overall, bleeding tendency was increased in the UFH group but this was not statistically significant.

Change in platelet count

The Li 2012 trial did not directly compare the platelet counts of the LMWH and UFH groups. However, the platelet count significantly decreased postoperatively in the UFH group (preoperative 278.4 ± 18.7 x 109/L, postoperative 194.3 ± 26.5 x 109/L; P < 0.05) but not in the LMWH group (preoperative 260.8 ± 32.5 x 109/L, postoperative 252.4 ± 29.1 x 109/L; P > 0.05).

Coagulation abnormalities

Chen 2001 monitored coagulability changes before anticoagulation and one hour, three days, and seven days after replantation. The parameters measured before treatment, after LMWH, and after UFH treatment were: antithrombin activity (0.97; 1.22; 1.56), factor Xa activity (1.02; 0.37; 0.58), bleeding time (2 min 24 sec; 3 min 18 sec; 4 min 35 sec), clotting time (8 min 55 sec; 12 min 20 sec; 25 min 32 sec), activated partial thromboplastin time (aPTT) (41 sec; 69 sec; 85 sec), and fibrinogen degradation product (FDP) concentration test (2.6 mg/L; 11.3 mg/L; 18.8 mg/L). No comparison was made between the LMWH and UFH groups but all data consistently showed that coagulability reduced more in the UFH group than in the LMWH group (see Table 1). The authors reported the mean value of the coagulation test by pooling the data at one hour, three days, and seven days after surgery but standard deviations were not available; further statistical analysis was thus not performed. Li 2012 measured the aPTT and found no significant differences in aPTT between both groups on the first postoperative day (LMWH 28.4 ± 3.3 sec, UFH 29.5 ± 3.1 sec; P > 0.05). However, the difference became significant during the second to seventh postoperative days (aPTT on the seventh postoperative day: LMWH 30.2 ± 2.1 sec, UFH 32.5 ± 2.2 sec; P < 0.05) (Analysis 1.4). It must be noted that aPTT changes were expected in the case of UFH, if given at large doses (such as in the Li 2012 trial), as raised aPTT levels are a therapeutic target for UFH but not for LMWH.

Table 1. Coagulability measures (Chen 2001)
  1. aPTT: activated partial thromboplastin time
    FDP: fibrinogen degradation product
    LMWH: low molecular weight heparin
    UFH: unfractionated heparin

Coagulability measuresbefore treatmentafter LMWH treatmentafter UFH treatment
Antithrombin activity0.971.221.56
Factor Xa activity1.020.370.58
Bleeding time2 min 24 sec3 min 18 sec4 min 35 sec
Clotting time8 min 55 sec12 min 20 sec25 min 32 sec
aPTT41 sec69 sec85 sec
FDP concentration test2.6 mg/l11.3 mg/l18.8 mg/l

Discussion

Summary of main results

The objective of this review was to compare LMWH with other anticoagulants or no treatment in terms of efficacy and safety in preventing microvascular occlusion after digital replantation. We found only two relevant RCTs (Chen 2001; Li 2012) comparing LMWH and UFH. In both small trials, the success rate of digital replantation did not significantly differ between the LMWH and UFH groups. Both trials found that LMWH was associated with less heparin-related hypocoagulability or reductions in the platelet count, and Chen 2001 observed an increased bleeding tendency in the UFH group. No trials comparing LMWH to other anticoagulants or no treatment were found. It remains unclear whether routine administration of anticoagulants improves the survival rate of replanted digits by preventing microvascular thrombosis. Compared to other anticoagulants, whether LMWH improves the success rate of replantation or has a reduced incidence of adverse effects is also unknown.

Overall completeness and applicability of evidence

Only two trials were identified for inclusion in this review. Both trials compared LMWH with UFH. No studies comparing LMWH with placebo, no treatment or other anticoagulants were identified. In addition, the two identified trials (Chen 2001; Li 2012) provide very limited evidence on the efficacy of LMWH after digital replantation. Therefore, most of the review question has not yet been resolved.

Although the success rate did not differ between the LMWH and UFH groups in both trials, the authors did not report whether there was any re-exploration and successful salvage of compromised digital perfusion. In those participants who failed replantation, the direct and indirect causes of microvascular occlusion were not reported. Occlusions of the repaired vessels may occur in either arteries or veins, or both. When the arterial inflow is blocked, the replanted digits will turn pale and cool, with decreased skin turgor and sluggish capillary refill. When the venous drainage is blocked, the replanted digits will become congestive and cyanotic. Although the Chen 2001 trial reported the incidence of arterial and venous insufficiency separately, a clear definition of these clinical conditions was not provided. The Chen 2001 trial also reported fewer bleeding events in the LMWH group compared to the UFH group, but the diagnostic criteria were not explicitly stated. This makes the judgement of the actual efficacy and complication rate of LMWH difficult.

We expected that subcutaneous LMWH or other anticoagulants might increase the patency rate of microvascular anastomosis performed on small sized, low-flow and severely injured vessels such as in the repair of digital veins as well as in replanting crushed or avulsed digits or fingertips. However, the available trials did not provide enough data to investigate all relevant types of injuries and the surgical outcomes.

Quality of the evidence

This review included only two RCTs with high or moderate risk of bias based on the various assessed parameters of methodological quality. Blinding and selective reporting (Chen 2001) and other bias (Li 2012) were identified as reasons for a high risk of bias in the included studies. A total of 114 participants with at least 122 completely or incompletely amputated digits, three palms, two forearms and three toes were included. Detailed information about the mechanism and level of amputation was unavailable. Chen 2001 compared the treatment outcomes of microvascular repair at different levels (digit, palm, forearm and toe), which might be a potential source of bias. For each single replanted digit, Li 2012 stated that the number of vessels that were repaired ranged from one artery and one vein to two arteries and three veins. This variation in surgical interventions might be another potential source of bias since complete occlusion of one vessel could have resulted in different outcomes if the number of repaired vessels was different.

Potential biases in the review process

We conducted the review according to the published protocol. We tried to avoid publication bias by searching a wide range of databases including two databases in Mandarin (CNKI and CEPS). However, it is still possible that some relevant trials published in different languages may not have been identified. Only two RCTs were identified and an objective assessment of publication bias was not performed. We could not obtain additional data from the trialists of the included studies since their contact information was not available. Outcomes from replantation of three palms and two forearms were included in this review because Chen 2001 pooled these results with the results from digital replantation and they could not be extracted. However, since this trial was a RCT, we assumed these five amputations (5/54) were randomly distributed and were evenly represented in the LMWH and UFH groups. We were unable to establish contact with the trialists to confirm this.

The type of study included in this review was only the RCT. It should be kept in mind that the systematic evaluation of adverse effects of LMWH or other anticoagulants may require other types of studies, such as cohort and case-control studies (Chi 2009; Cochrane Adverse Effects Methods Group 2012; Loke 2007).

Agreements and disagreements with other studies or reviews

Currently there is little evidence to guide perioperative anticoagulation in digital replantation. The best available evidence consists mainly of retrospective studies (Buckley 2011; Fukui 1989; Fukui 1994; Han 2000; Niibayashi 2000; Noguchi 1999; Oufquir 2006; Poole 1977; Rapoport 1985; Veravuthipakorn 2004; Vretos 1995), comparative studies (Azolov 1983; Furnas 1992; Gao 2007; Maeda 1991; Nikolis 2011; Yang 2008; Yu 2012; Zhang 2002; Zhang 2004), prospective cohort studies (Loisel 2010), or animal studies (Rooks 1994). Most of these studies compared anticoagulants other than LMWH, and many of the authors included major limb replantation or free flap transfer in their studies (Askari 2006; Azolov 1983; Fukui 1994; Khouri 1998; Maeda 1991; Pederson 2008). One RCT compared UFH with no treatment (Shu 2011). In the Shu 2011 trial, 24 patients with 42 digital replantations were randomised to the UFH group (50 IU/kg IV bolus) versus 25 patients with 35 digital replantations randomised to the control group. A per protocol analysis yielded three versus three arterial occlusions (RR 0.83; 95% CI 0.18 to 3.87), two versus one venous occlusions (RR 1.67; 95% CI 0.16 to 17.62), and four versus three replantation failures (RR 1.11; 95% CI 0.27 to 4.63). The results showed that single-dose UFH had no positive effect on microvascular patency and the success rate of replantation. It is worth noting that 85.7% (36/42 in the UFH group and 30/35 in the control group) of the amputations were incomplete, and 72.7% (30/42 and 26/35 respectively) of the injury mechanisms were crushing and avulsion. To date, no other systematically reviewed evidence concerning LMWH and digital replantation can be found.

The efficacy of antithrombotic agents for preventing thrombosis had been reviewed systematically for patients that received infrainguinal arterial bypass surgery (Geraghty 2011). Three trials comparing LMWH to UFH were included (Norgren 2004; Samama 1994; Swedenborg 1996). Meta-analysis of 507 participants (Norgren 2004; Samama 1994) failed to demonstrate a significant difference on early patency, and only marginally favoured LMWH versus UFH for early graft thrombosis on day 30 (odds ratio (OR) 0.54; 95% CI 0.33 to 0.90). Another included trial (Jivegard 2005) compared the primary graft patency in 229 patients. One hundred and sixteen participants were randomised to receive LMWH (dalteparin 5000 IU SC once daily for three months) versus 113 that received placebo; all participants received aspirin. The bypass occlusions at one month (8 versus 11; OR 0.69; 95% CI 0.27 to 1.78), three months (20 versus 25; OR 0.75; 95% CI 0.39 to 1.44), and 12 months (44 versus 46; OR 0.97; 95% CI 0.56 to 1.69) revealed no significant differences between the groups at any of these time points. One included trial in the Geraghty 2011 review (Edmondson 1994) compared LMWH with acetylsalicylic acid and dipyridamole (ASA and DIP). Ninety-four patients were randomised to receive 2500 IU SC LMWH (Fragmin) once daily, and 106 patients to receive 300 mg aspirin and 100 mg dipyridamole, both three times per day for three months. Bypass occlusions at six months (12 versus 30; OR 0.39; 95% CI 0.20 to 0.78) and 12 months (21 versus 38; OR 0.52; 95% CI 0.29 to 0.96) showed a significant positive effect for LMWH. Further subgroup analysis found that this positive effect was only significant in patients undergoing operations for limb salvage and not in patients having operations for claudication. It remains unclear whether the efficacy of LMWH in maintaining the patency of large-sized vessels is consistent with the findings when treating much smaller sized vessels, such as digital arteries and veins.

Authors' conclusions

Implications for practice

The current limited evidence showed no significant differences in the efficacy of LMWH and UFH in preventing microvascular occlusion after digital replantation, but LMWH was associated with a lower risk of postoperative bleeding and hypocoagulability. There are many unclear risks of bias, which make the quality of evidence uncertain.

Implications for research

No placebo-controlled trials were identified by this systematic review. Further double-blind RCTs should be conducted to compare LMWH with placebo and to evaluate the efficacy of LMWH in reducing the risk of anastomotic thrombosis. Detailed information regarding the size and quality of the involved vessels and the interventions used should be reported for subgroup analysis in order to limit the risk of bias and to provide conclusive evidence. Prospective cohort studies and case-control studies would also be useful in assessing the adverse effect of LMWH and should be included in future reviews. Further reviews that include all anticoagulants and their application in replantation or microsurgery are warranted. A cost-effectiveness analysis of these interventions should also be considered.

Acknowledgements

We acknowledge the Cochrane Peripheral Vascular Diseases Review Group and Marlene Stewart for their assistance in the development of this review.

Data and analyses

Download statistical data

Comparison 1. LMWH versus UFH
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Success rate of replantation2 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
1.1 Unit of analysis: numbers of participants1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
1.2 Unit of analysis: numbers of digits1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2 Compromised microcirculation1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
2.1 Arterial1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.2 Venous1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3 Haemorrhage1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
3.1 wound bleeding1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.2 hematuria1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.3 gingival bleeding1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.4 fecal occult blood1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.5 nasal bleeding1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.6 ecchymosis1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
4 aPTT1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
Analysis 1.1.

Comparison 1 LMWH versus UFH, Outcome 1 Success rate of replantation.

Analysis 1.2.

Comparison 1 LMWH versus UFH, Outcome 2 Compromised microcirculation.

Analysis 1.3.

Comparison 1 LMWH versus UFH, Outcome 3 Haemorrhage.

Analysis 1.4.

Comparison 1 LMWH versus UFH, Outcome 4 aPTT.

Appendices

Appendix 1. CENTRAL search strategy

#1 MeSH descriptor: [Hand] explode all trees 1847

#2 MeSH descriptor: [Finger Injuries] this term only 84

#3 hand 14816

#4 finger 2728

#5 thumb 548

#6 digit 1931

#7 (radial or digital or pollicis or ulnar or palmar) near/4 arter* 912

#8 MeSH descriptor: [Radial Artery] this term only and with qualifiers: [Blood supply - BS, Surgery - SU] 81

#9 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 19219

#10 MeSH descriptor: [Amputation] explode all trees 297

#11 amput* 1237

#12 avuls* 179

#13 mutilat* 112

#14 MeSH descriptor: [Ischemia] explode all trees 775

#15 ischaemi* 4842

#16 ischemi* 12915

#17 MeSH descriptor: [Replantation] explode all trees 37

#18 MeSH descriptor: [Reconstructive Surgical Procedures] this term only 494

#19 replant* or *attach* 4439

#20 revascular* 4562

#21 salvag* 1734

#22 anastomosis 1643

#23 reconstruct* or re-construct* 4158

#24 re-vascula* 20

#25 MeSH descriptor: [Microsurgery] this term only 354

#26 microsurg* 609

#27 micro-surg* 17

#28 #10 or #11 or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23 or #24 or #25 or #26 or #27 31500

#29 MeSH descriptor: [Heparin, Low-Molecular-Weight] explode all trees 1608

#30 heparin* 7684

#31 LMWH 823

#32 nadroparin* or fraxiparin* or enoxaparin 1380

#33 Clexane or klexane or lovenox 77

#34 dalteparin or Fragmin or ardeparin 582

#35 normiflo or tinzaparin or logiparin 226

#36 Innohep or certoparin or sandoparin or reviparin 182

#37 clivarin* or danaproid or danaparoid 94

#38 antixarin or ardeparin* or bemiparin* 63

#39 Zibor or cy 222 or embolex or monoembolex 73

#40 parnaparin* or rd 11885 or RD1185 41

#41 tedelparin or Kabi-2165 or Kabi 2165 68

#42 emt-966 or emt-967 or pk-10 169 or pk-10169 or pk10169 19

#43 fr-860 or cy-216 or cy216 80

#44 seleparin* or tedegliparin or seleparin* or tedegliparin* 12

#45 wy90493 or "wy 90493" 9

#46 ("kb 101" or kb101 or lomoparan or orgaran) 64

#47 parnaparin or fluxum or lohepa or lowhepa 49

#48 op 2123 or parvoparin 13

#49 ave 502 63

#50 calciparin* 28

#51 #29 or #30 or #31 or #32 or #33 or #34 or #35 or #36 or #37 or #38 or #39 or #40 or #41 or #42 or #43 or #44 or #45 or #46 or #47 or #48 or #49 or #50 8304

#52 #9 and #28 and #51 11 in Trials

Appendix 2. Authors' PubMed search strategy

1. hand[MeSH Terms] OR hand[All Fields] OR hands[All Fields] 341314

2. fingers[MeSH Terms] OR fingers[All Fields] OR finger[All Fields] 83476

3. thumb[MeSH Terms] OR thumb[All Fields] OR thumbs[All Fields] 14272

4. digit[All Fields] OR digits[All Fields] OR digital[All Fields] 90873

5. radial[All Fields] OR digital[All Fields] OR pollicis[All Fields] OR ulnar[All Fields] OR palmar[All Fields] AND Arter* 14401

6. ((((#1) OR #2) OR #3) OR #4) OR #5 469579

7. amput* 37758

8. avuls* 8871

9. mutilat* 6659

10. ischaemia[All Fields] OR ischemia[MeSH Terms] OR ischemia[All Fields] OR ischemic[All Fields] 276229

11. replantation[MeSH Terms] OR replantation[All Fields] OR replantat* 7276

12. revascular* 43186

13. salvage* 35015

14. microsurgery[MeSH Terms] OR microsurgery[All Fields] OR microsurg* 35573

15. Reconstruct* 198775

16. ((((((((#7) OR #8) OR #9) OR #10) OR #11) OR #12) OR #13) OR #14) OR #15 592309

17. heparin[MeSH Terms] OR heparin[All Fields] OR heparin* 86953

18. anticoagulants[MeSH Terms] OR anticoagulants[All Fields] OR anticoagulant[All Fields] OR anticoagulants[Pharmacological Action] OR anticoagula* 206013

19. antithrombo* 12404

20. UFH OR UH OR LMWH 15916

21. nadroparin* OR fraxiparin* OR enoxaparin 3957

22. Clexane OR klexane OR lovenox 3444

23. dalteparin OR Fragmin OR ardeparin 1226

24. normiflo OR tinzaparin OR logiparin 395

25. Innohep OR certoparin OR sandoparin OR reviparin 533

26. clivarin* OR danaproid OR danaparoid 589

27. antixarin OR ardeparin* OR bemiparin* 108

28. Zibor OR cy 222 OR embolex OR monoembolex 216

29. parnaparin* OR rd 11885 OR RD1185 44

30. tedelparin OR Kabi-2165 OR Kabi 2165

31. emt-966 OR emt-967 OR pk-10 169 OR pk-10169 OR pk10169 1048

32. fr-860 OR cy-216 OR cy216 1555

33. seleparin* OR tedegliparin OR seleparin* OR tedegliparin* 1010

34. wy90493 OR wy 90493 0

35. kb 101 OR kb101 OR lomoparan OR orgaran 1022

36. parnaparin OR fluxum OR lohepa OR lowhepa 54

37. op 2123 OR parvoparin 46

38. ave5026 4

39. calciparin* 53

40. ((((((((((((((((((((((#17) OR #18) OR #19) OR #20) OR #21) OR #22) OR #23) OR #24) OR #25) OR #26) OR #27) OR #28) OR #29) OR #30) OR #31) OR #32) OR #33) OR #34) OR #35) OR #36) OR #37) OR #38) OR #39 240139

41. ((#6) AND #16) AND #40 654

Appendix 3. Authors' CNKI search strategy

(((題名= 低分子肝素 OR 關鍵詞 = 低分子肝素 OR 摘要 = 低分子肝素) OR (題名= 低分子量肝素 OR 關鍵詞 = 低分子量肝素 OR 摘要 = 低分子量肝素)) AND ((題名 =斷指 OR 關鍵詞 =斷指 OR 摘要 =斷指 OR 全文 =斷指) OR (題名 = 再植 OR 關鍵詞 = 再植 OR 摘要 = 再植 OR 全文 = 再植))) 32

Appendix 4. Authors' CEPS search strategy

(肝素 AND 斷指)=所有欄位 5

Contributions of authors

Yi-Chieh Chen (YCC), Fuan Chiang Chan (FCC), and Yu-Wen Wen (YWW) wrote the protocol. YCC and FCC selected trials, assessed trial quality, and extracted the data. YCC and Ching-Chi Chi (CCC) interpreted the data and conducted statistical analyses. YCC drafted the final review with contributions from CCC.

Declarations of interest

None known

Sources of support

Internal sources

  • No sources of support supplied

External sources

  • Chief Scientist Office, Scottish Government Health Directorates, The Scottish Government, UK.

    The PVD Group editorial base is supported by the Chief Scientist Office.

Differences between protocol and review

Outcomes from replantation of three palms and two forearms were included in this review because Chen 2001 pooled these results with the results from digital replantation and could not be extracted. However, since this trial was a RCT, we assumed these five amputations were randomly distributed evenly in the LMWH and UFH groups. We were unable to establish contact with the trialists to confirm this.

One secondary outcome (coagulability assessed by antithrombin activity, factor Xa activity, bleeding time, clotting time, aPTT or FDP concentration test) that was not prespecified in the protocol was included in the review as it was judged to be clinically important. Otherwise no other difference exists between protocol and review.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Chen 2001

Methods

Design: RCT

Randomisation: method unknown

Blinding: nil
Intention-to-treat/Loss to follow up: unclear. The authors did not report if some patients were randomised but did not complete the trial.

Participants

Setting: China Railway Group's four innings second hospital, Fuyang City, Anhui Province, China

Study period: 1999 Oct - 2000 Jun

54 participants of replantation including 39 single digits, 7 multiple digits, 3 palms, 2 forearms and 3 toes were randomised into two groups: group A (LMWH) aged 4.5 - 57 yrs, mean 25.8, group B (UFH) aged 8-55 yrs, mean 26.6 yrs
Inclusion criteria:

- no history of major bleeding disorders

- normal bleeding time and clotting time before replantation
Exclusion criteria: not reported

Interventions2 intervention groups
Treatment (group A , n = 26): livaracine (Lee's Pharmaceutical Holdings Ltd, Hong Kong, China) 5000 IU SC 30 min before surgery, followed by 2500-5000 IU SC q12h x 7 days
Control (group B, n = 28): UFH 2500 IU IV 0.5 - 1 hr before surgery, followed by 1250 IU q12h x 7 -10 days; also received low molecular weight dextran 500ml twice a day
Outcomes

1. Success rate of replantation: LMWH = 24/26 (89.2%), UFH = 25/28 (92.3%), P > 0.05
2. Bleeding tendency:

- LMWH group: 3 wound bleeding, 1 ecchymosis, 1 haematuria

- UFH group: 5 wound bleeding, 3 ecchymosis, 2 nasal bleeding, 3 gingival bleeding, 2 haematuria, 1 faecal occult blood

3. Coagulation abnormality (assessed by antithrombin activity, factor Xa activity, bleeding time, clotting time, aPTT, FDP concentration test): UFH > LMWH (see also Table 1)

NotesThe study authors reported that success rate was calculated by individual patients; the mean value of coagulation test was obtained by pooling the data of 1hr, three days and seven days after surgery; statistical evaluation was not reported
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote p. 53: "...patients were randomised into group A and B... (author's translation)"

Comment: Method not described

Allocation concealment (selection bias)Unclear riskComment: Method of allocation not described
Blinding of participants and personnel (performance bias)
All outcomes
High riskComment: LMWH was administered subcutaneously but UFH was administered intravenously
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskComment: No description of blinding of outcome assessment
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskComment: No description of dropouts or withdrawals
Selective reporting (reporting bias)High riskComment: Standard deviation of aPTT, bleeding time and FDP (mg/L) were not reported, platelet count change not reported
Other biasUnclear riskComment: Outcomes of digital amputation (n = 46) were mixed with amputation of palm (n = 3), forearm (n = 2) and toe (n = 3) and cannot be extracted. The mean value of coagulation test was obtained by pooling the data of 1h, 3d and 7d; SD value was not available; statistic evaluation was not reported. Low molecular weight dextran was used only in the control group, which might have biased the results. The definition or arterial and venous insufficiency was not described

Li 2012

  1. a

    aPTT: activated partial thromboplastin time
    FDP: fibrinogen degradation product
    IU: international unit
    IV: intravenous
    LMWH: low molecular weight heparin
    po: per os; by mouth, orally
    qd: quaque die; every day, once daily
    RCT: randomised controlled trial
    SC: subcutaneous
    tid: ter in die; three times a day
    UFH: unfractionated heparin
    yrs: years

Methods

Design: RCT

Randomisation: method unknown
Blind assessment: double blind
Intention-to-treat/loss to follow-up: unclear. The authors did not report if some patients were randomised but did not complete the trial

Participants

Setting: Integrative Medicine Hospital of Guangdong Province, Foshan City, China

Study period: since 2006
60 participants (46 male, 14 female) aged 17- 43 yrs (mean 24.5 yrs) with 69 complete or incomplete amputated digits
Inclusion criteria: amputation of digit(s) between the level of metacarpophalangeal joint and distal phalangeal base, non-salvageable without vascular anastomosis
Exclusion criteria: not stated

Interventions2 intervention groups
Treatment (30 participants with 35 replanted digits): LMWH (exact type unknown) 0.4 ml SC q12h x 7-10 days
Control (30 participants with 34 replanted digits): UFH 10000 IU SC q12h x 7-10 days
Both groups also received:
- papaverine 60 mg/IM/q6h x 7-10 days
- low molecular weight dextran 500ml IV q12h x 7-10 days
- dipyridamole 25mg po tid x 7-10 days
- aspirin 50mg po qd x 7-10 days
Outcomes1. Success rate of replantation:
LMWH = 33/35 (94.3%), UFH = 32/34 (94.15), P > 0.05
2. aPTT prolongation: UFH > LMWH (between 2nd-7th postoperative days), P < 0.05
3. Platelet count: LMWH unchanged, UFH dropped significantly than before surgery
NotesSuccess rate was calculated by digits
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote p. 205: "...participants were randomised into two groups... (author's translation)"

Comment: Method not described

Allocation concealment (selection bias)Unclear riskComment: Method of allocation not described
Blinding of participants and personnel (performance bias)
All outcomes
Low risk

Quote p. 205: "...double blind principle was followed... (author's translation)"

Comment: The outcome is not likely to be influenced

Blinding of outcome assessment (detection bias)
All outcomes
Low risk

Quote p. 205: "...double blinding principle was followed... (author's translation)"

Comment: Outcome measurement is not likely to be influenced

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskComment: No description of dropouts or withdrawals
Selective reporting (reporting bias)Unclear riskComment: The trialists did not report the incidence of compromised microcirculation requiring surgical re-exploration or re-anastomosis. The direct or indirect cause of microvascular insufficiency was also not provided
Other biasHigh riskComment: The LMWH used in this trial was not reported. The dosage of LMWH was not reported; only the volumetric dosage was provided. The specific mechanisms of injury leading to amputation and the severity of soft tissue damage between each group were not mentioned. In some replanted digits only one digital artery and one digital vein were repaired, while in some other cases two digital arteries and three digital veins were repaired. Occlusion of a repaired vessel might result in different outcome if the number of repaired vessels was different

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    IA: intra-arterial
    IM: intramuscular
    IV: intravenous
    LMWH: low molecular weight heparin
    SC: subcutaneous
    U: unit
    UFH: unfractionated heparin

Azolov 1983Comparative study, compared coumarin, nitrofarin, aspirin and dextrans; not focused on LMWH [Article in Russian]
Buckley 2011Case report and review of evidence
Fukui 1989Case series, 13 cases (replantations and damaged digital arteries) received daily intra-arterial infusion of (1) 240,000 U of urokinase, (2) 40 micrograms of prostaglandin E1, (3) 10,000 U (maximum) of heparin, and (4) low molecular weight dextran for 10 consecutive days
Fukui 1994Retrospective survey conducted among 94 members of the Japanese Society of Reconstructive Microsurgery on the present status of replantation in Japan and included 9,664 replanted extremities (157 upper arms, 415 forearms, 471 hands, 8,320 digits, 33 thighs, 103 calves, 37 feet, and 128 toes). Outcomes of continuous local intra-arterial infusion of various anticoagulant and thrombolytic agents at Nara Medical University were reviewed
Furnas 1992Comparative study, compared the incidence of blood loss and blood transfusion between patients with replanted digits receiving one anticoagulant in addition to aspirin (n = 40) and two or more anticoagulant in addition to aspirin (n = 15)
Gao 2007Comparative study, compared the incidence of vascular insufficiency after digital replantation between two groups: the experimental group (n = 76) received clexane 40 mg SC daily as well as low molecular weight dextran and papaverine, the control group receive the same treatment but clexane [Article in Chinese]
Han 2000Retrospective study, used topical and systemic anticoagulation to maintain external bleeding for fingertip replantation to overcome venous congestion in 140 cases without venous repair or in cases in which there is obstruction of a repaired vein
Loisel 2010Prospective study without control group, enrolled 13 cases of replantation (one devascularisation of hand, two complete amputations of hand, four ring fingers and six complete amputations of finger) using local irrigation of anastomosis with urokinase and LMWH to reduce the thrombosis rate [Article in French]
Maeda 1991Comparative study, compared intra-arterial, intravenous infusion of urokinase, heparin and PGE1 in 44 patients. IA group included 24 replantation (17 digit, 4 hand, one forearm and 2 toe replantations) and 5 open fracture cases; IV group included 10 replantations; control group were 5 patients with hand injuries
Niibayashi 2000Retrospective study, 252 consecutive digital replantations and revascularisations in 201 patients were reviewed to determine factors influencing survival, including the application of anticoagulant agents
Nikolis 2011Historical control study, the retrospective review group included 175 digits (104 revascularisations and 71 amputations), IV heparin was used in 35.1% of the cases; the prospective group included 106 digits (45 revascularisations and 61 amputations), IV heparin was used in 14.6% of the cases. The two cohorts were compared with regards to demographics, injury characteristics, postoperative thromboprophylaxis medication as well as complication and success rates
Oufquir 2006Retrospective study, 15 amputated digits developed "no reflow phenomenon" after replantation and received intra-arterial infusion of antithrombotic and fibrinolytic agents [Article in French]
Pomerance 1997Retrospective study, 127 digits and 24 thumbs in 106 patients were reviewed to assess early results and complications, and to determine if a short course (minimum 3 days) of dextran-40 anticoagulation adversely affected digit survival
Rooks 1994Double-blind animal study, used a rat femoral artery crush-avulsion model to analyse the efficacy of intra-arterially delivered anticoagulants against similar systemically administered intra-venous anticoagulants
Shu 2011Randomised controlled trial, 49 patients were randomised to receive single dose IV UFH (50 IU/kg) (24 participant with 42 digits), or no treatment (25 participant with 35 digits) [Article in Chinese]
Vretos 1995Retrospective study. 43 patients received replantation, revascularization or free flap transfer. Antithrombotic medication included (1) aspirin and dipyridamole; (2) low molecular weight dextran; (3) streptokinase and streptodonase; (4) diclofenac. No control group
Yang 2008Historical control study, 40 patients with 56 amputated digits received replantation and fasudil treatment were included as the experimental group; comparison was made with retrospective data reviewed from 32 patients with 45 digital replantation who received postoperative LMWH and papaverine as the control group
Yu 2012Comparative study, 41 cases with 44 amputated digits in the control group received IM injection of heparin and papaverin after replantation; another 41 cases with 44 digital replantation received continuous local infusion of solution with heparin, lidocaine and dexamethasone via a catheter inserted into the flexor tendon sheath [Article in Chinese]
Zhang 2002Comparative study, compared the efficacy and adverse effect between two groups: 12 cases with 17 amputated digits received LMWH, the other 12 cases with 19 digits received UFH [Article in Chinese]
Zhang 2004Comparative study, included 45 cases with 60 digital amputations which were caused by avulsion or crush injury or happened on the fingertips. Twenty-three cases received postoperative LMWH, the other 22 cases received UFH

Ancillary