The fundamental goals of treatment of an acute ATR are to restore length and tension of the tendon and thus to optimize a patient's ability to return to the previous level of activity, with as few complications as possible. In complete ATR there always come to the retraction of triceps muscle, resulting in a gap, which is filled in a healing process with fibrous tissue, that is but not as strong as tendon [30, 55, 60, 85]. Restoration of the length with a good approximation of the torn ends is believed to have an impact on the low number of re-ruptures in operative treatment [30, 55, 60, 85]. In conservative treatment the extent of the fibrous tissue could somehow be influenced by positioning of the foot in PF and with early functional treatment [5, 9–13, 85, 86]. Early ankle motion exercises and weightbearing might have an impact on the histologic properties of a healing tendon (tissue arrangement and collagen type) and thus on stiffness, adhesion formation, gapping resistance and (re)rupture strength [85–87]. The proposed mechanism of this is that physiological forces promote gene expression of type I collagen formation during healing and that tension causes the collagen to be deposited and aligned in parallel fashion [88]. The same beneficial effect of faster healing of tendons if they are subjected to loading can be found after operative treatment as well [86, 87]. It has been noted but that disproportionate weakness in end-range plantar flexion, decreased passive stiffness in dorsiflexion and inability to perform a decline heel rise (lower strength) are evident after ATR and repair [89]. Possible causes include anatomical lengthening, increased tendon compliance and insufficient tendon rehabilitation after ATR [50, 89] or could be chronic adaptations associated with ATR [85].
Stronger repair with good adaptation of the torn ends could thus be beneficial in any type of repair, including percutaneous repair of the ruptured AT. This type of repair has been criticized to be weaker than open operative repair [6, 51, 56–58] and that in closed technique the ends cannot be visualized and brought into a good, completely approximated position, with difficulties achieving appropriate length and tension at the repair site [30, 57, 58, 73]. Semi-open or minimally invasive (MI) methods of AT repair were thus proposed, many using special (costly) instruments (like Achillon® system (Integra Life Sciences Corporation, Plainsboro, NJ), PARS™ system (Arthrex, Naples, FL), Achilles midsubstance Speed Bridge repair variation (Arthrex, Naples, FL)) with several sutures or anchors [23–28, 52]. The torn ends in ruptured AT are often friable and uneven and pulling them together to achieve a good approximation has proven difficult [20, 24, 25, 30, 60]. The majority of the proposed percutaneous techniques enable pulling the torn ends only at one side, what might leave the gap on the opposite side (so called »fish - tail« effect) or causes tearing and cutting a tendon with the thread during stronger pulling in an attempt to reapproximate both sides [30, 60]. This is why we recommend more »rough« material, like Vicryl instead of for instance more »smooth« PDS® No.2 (polydioxanone) (Ethicon, Inc., a division of Johnson&Johnson, Sommerville, NJ), with a greater risk of cutting a tendon. In the types of »crossing technique« [90, 91], interference with the approximation at the site of the rupture, can occur. This results in a potential residual gap that is healed with weaker fibrous tissue and with lengthening of the tendon. The proposed modified technique is one of the first percutaneous methods performed under local anesthesia [21] and at the moment the only closed one that enables approximation of torn ends by pulling them symmetrically and simultaneously at both sides, using the principle of »double pulley technique« [30, 60]. Despite this technique is more demanding, it enables less force needed to make a good apposition of the torn ends, with less risk of cutting the tendon during pulling and minimizes the pullout forces at the junction of the suture-tendon interface [30, 60, 92]. Instead of opening the site of rupture to control the approximation of the torn ends, what is used in mini invasive methods, approximation can be controlled and assisted by ultrasonography [30, 60], which we recommend during initial use of the proposed method and until the surgeon is comfortable with the technique and proper pulling strength. It is very important to approximate the torn ends until the defect is no longer palpable and visible by ultrasonography and maximal PF of the foot during reapproximation and tightening probably assists in this maneuver. It seems therefore that in the term of controlling apposition of the ends, there is no real need to open this area or to use arthroscope, particularly if keeping in mind losing biological potential and increasing the risk of infection.
Resistance to elongation together with resistance against gapping and failure under cyclic (repetitive) loading was intensively studied in biomechanical testing of several methods, including open, semi-open (minimally invasive) and percutaneous (closed) types of repair [51, 59, 61–74]. The results of biomechanical (cadaveric) testing of percutaneous and mini-invasive methods are listed in Table 5.
Besides the technique itself, the strength of tendon repair depends on three factors: the holding capacity of the suture material within the tendon (the coefficient of friction); the strength of the knot; and the strength of the suture material itself [73]. More strands and knots with locking type of sutures do have impact on strength of the repair, but also on biological potential of healing (diminished tendon blood supply, additionally if performed in (semi)open way) and possible complication (particularly with the use of non-resorbable thread) like adhesions, suture extrusion and friction within the paratenon with affecting glide, that might end with the need for reoperation [28, 93]. The proposed method showed in biomechanical testing the highest strength among closed percutaneous methods. Semi-open methods with previously cited tendon repair systems showed in some (but not all) biomechanical testing stronger repair in comparison to the proposed method [51, 59, 61–74], but demand special instruments and more (generally 6) sutures (with anchors) and produce higher costs with the potential risks as stated above. The strongest repair in biomechanical testing showed otherwise open repair with Krackow locking loop (582 N) [94], but if instead of No.2 polyblend sutures (Fiberwire, Arthrex, Naples, Fl, USA) No.1 Ethibond suture (Ethicon, Inc., a division of Johnson&Johnson, Sommerville, New Jersey, USA) was used, the load to failure with the same type of repair was only 147 N [63] or 161 N [95].
Table 5
Comparison of Achilles tendon repairs ranked by load to failure
Technique (closed)
|
Suture material
|
No. of strands
|
Load to failure (N)
|
Author
|
Kessler (type)
|
2 − 0 Fiberwire
|
2
|
38
|
61
|
Bunnell (type)
|
0 Ticron
|
2
|
78
|
62
|
Bunnell (type)
|
1 Ethibond
|
2
|
93
|
63
|
Carmont-Maffulli
|
2 Prolene
|
4
|
106
|
64
|
Ma-Griffith
|
2 Vicryl
|
2
|
111
|
59
|
Kessler (”Dresden”)
|
1 PDS
|
2
|
137
|
65
|
Double Kessler (type)
|
1 − 0 Silk
|
4
|
154
|
66
|
Bone Anchor repair
|
1 Panacryl
|
4
|
166
|
67
|
Kessler (”Dresden”)
|
2 Mersilene
|
2
|
167
|
68
|
Bone Anchor repair
|
1 PDS-II
|
4
|
185
|
67
|
Calcaneal Tunnel
|
1 Panacryl
|
4
|
186
|
67
|
Calcaneal Tunnel
|
1 PDS-II
|
4
|
195
|
67
|
Modified (Cretnik)
|
2 Vicryl
|
4
|
214
|
59
|
Technique (semi-open (MI))
|
Suture material
|
No. of strands
|
Load to failure (N)
|
Author
|
Kessler (type)
|
1 Ethibond
|
2
|
85
|
63
|
Achillon
|
2 Ethibond
|
6
|
100
|
51
|
Achillon
|
2 Prolene
|
6
|
104
|
64
|
Kessler (type)
|
2 Ticron
|
2
|
123
|
69
|
Kessler (”Dresden” (modif.))
|
1 PDS
|
2
|
137
|
65
|
Bunnell (type)(”Majewski”)
|
1 PDS
|
2
|
139
|
65
|
Achillon
|
2 Ticron
|
6
|
153
|
69
|
Double Kessler (”Webb-Bannister”)
|
2 Mersilene
|
4
|
166
|
68
|
Kessler (type)
|
0.7 mm PDS (cord)
|
2
|
193
|
70
|
PARS
|
2 FiberWire
|
6 (sutures)
|
206
|
71
|
Bunnell (type)
|
0.7 mm PDS (cord)
|
2
|
255
|
70
|
Achillon
|
2 FiberWire
|
6
|
299
|
72
|
Achillon
|
1 Ethibond
|
6
|
342
|
73
|
PARS
|
2 FiberWire
|
6 (sutures)
|
353
|
74
|
PARS (”Speed Bridge”)
|
2 FiberWire
|
6 (anchors)
|
385
|
71
|
PARS (”Speed Bridge”)
|
2 FiberWire
|
6 (anchors)
|
385
|
72
|
Stronger repair could support more confidence and resistance against widening of the gap during the functional rehabilitation and healing with elongation of the ruptured AT [51–54, 59]. These all might be beneficial also in reducing the incidence of re-rupture. Percutaneous repair is associated with a re-rupture rate from 2.6–16.7% [22, 26, 30, 39, 56–58, 96, 97], although some papers about percutaneous repair without this complication can be found [20, 21, 25, 28]. Hsu et al. reported in their retrospective cohort study no re-rupture in treatment of 101 patients with PARS™ system and 169 patients in an open way [28]. Bartel et al. reported in the systematic review of incidence of complications after Achillon® system in 8 eligible of 33 studies the re-rupture incidence of 3.2% (in 8 of 253 patients) [98]. Yang et al. found in their meta-analysis of 5 randomized controlled trials and 7 retrospective cohort studies about outcomes and complications of percutaneous versus open repair of acute ATR involving 815 patients, 13 (3.1%) out of 424 percutaneously treated patients who experienced re-rupture [99]. In our series there was 1 patient (1.61%) in IG out of 62 procedures who suffered a re-rupture. As none of the patients in the FG experienced a re-rupture, the proposed modified method showed in our study strong and reliable enough for early functional treatment and with lower re-rupture rate even in comparison to the results in meta-analyses for the open operative reconstruction (1.4–4.4%) [35, 38–41, 43, 99, 100].
Re-rupture and (severe) wound infection are the most important complications with lasting negative effect on outcome [30, 39, 40, 43, 98, 101–103]. Open surgery around the AT has a wound-related complication rate of between 8.2% and 34.1%, [11, 39, 102], of which at least half are due to infection [103]. Meta-analyses in open procedures revealed deep infection rate of 2.3% [40, 43]. Reports of infection rate with the use of percutaneous and semi-open methods differ quite a lot from 0–13.3% [20, 28, 30, 39, 57, 60, 96, 104]. Bartel et al. reported in the systematic review of incidence of complications after Achillon® system the infection rate of 0.8% (in 2 of 253 patients) [98]. Saxena et al. reported that the overall incidence of wound infections in surgeries pertaining to the AT was 3.1% (in 7 of 219 patients) [103]. Yang et al. found in their meta-analysis 9 eligible studies revealing the occurrence rate of deep infection 0.6% with percutaneous treatment and 3.6% with open treatment in total [99]. A subgroup analysis of only 5 RCTs showed no significant difference between these two groups (RR = 0.42, 95% CI 0.11 to 0.96, p = 0.99; I2 = 0%) [99]. Grassi et al. found in their meta-analysis of 8 randomized controlled trials with 182 patients treated with minimally invasive surgery and 176 treated with open repair, significantly decreased risk ratio (RR) of 0.15 (95% confidence interval [CI] = 0.05 to 0.46, p = 0.0009) for wound infection after minimally invasive surgery [105]. When deep and superficial infections were analyzed separately, only the superficial infections remained significantly decreased in the minimally invasive surgery group after both random and fixed-effect meta-analysis with a relative risk reduction (RRR) of 83% [105]. In our series no patient received antibiotic prophylaxis, including patients with diabetes, corticosteroid use and smoking, that were found together with age as major risk factors for infection [98, 102, 103] and no infection or problems with wound healing occurred.
Percutaneous methods, particularly closed (blind) suturing, were criticized about high rate of sural nerve injury [6, 24, 56, 58]. Semi-open methods were introduced to reduce the problems with sural nerve entrapment as well as for the control of approximation of torn ends [23, 24, 26–28]. Indeed, opening the site of the rupture and introducing different types of (costly) devices or surgical instruments (forceps) can reduce the occurrence of sural nerve problems, but raises the same concerns as with the open fractures, with losing (biological) healing potential (mediators and molecules in hematoma) and increasing risk of infection and wound dehiscence [28, 30, 60, 98, 102]. There were attempts to solve these problems with biological enhancement, such as adding platelet rich plasma (PRP), but there is still lack of evidence to support this solution [104]. Some authors reported complete elimination of the sural nerve problems using semi-open methods [22, 26, 28]. According to meta-analysis, this problem occurs in 0.78% in conservative treatment [40] as well as in open operative repair in up to 8.76% of cases, despite being able to preserve the sural nerve through cautious operative technique [40, 106]. Assal et al. reported no sural problems in the first paper using Achillon® (semi-open) system [26]. Bartel et al. reported but in the systematic review of incidence of complications after using the same system 1,2% (in 3 of 253 patients) sural nerve injuries [98]. Aibinder et al. found in their cadaveric study using the Achillon device that 8 of the 54 needle passes (14.8%) directly pierced the substance of the sural nerve, what could be diminished with external rotation of device [107]. Majewski et al. proposed additional stab incisions on the lateral site to expose the sural nerve to avoid hitting it during repair, what otherwise occurred in 18% of their 84 percutaneously treated patients [108]. In our first series [21] with the use of the proposed method in 36 patients there were no problems with sural nerve entrapment. In our longer-term study [30] from 1991 to 1997, there were 6 (4.47%) out of 134 cases with disturbances of sensibility, that spontaneously resolved in 2 to 10 months without any surgical intervention needed. In this series in 3 (4.83%) of 62 cases problems (paresthesias) with sural nerve disturbances were noted. Interestingly no-one claimed about this problem immediately after the surgery (but after 1 to 3weeks) and no-one experienced anesthesia in the area of the sural nerve. If operating under local anesthesia, patients should be warned to immediately report if any changes in sensation occur within the area of anesthetic infiltration, particularly at the lateral site, where sural nerve crosses lateral border of the tendon (8.7 to 12.4 cm proximally to the AT insertion [77]) and where the position of the entry site of the thread shouldn't cross lateral edge of a tendon. In our series we used resorbable thread (Vicryl No.2) and there was no need to perform any surgical revision or release in patients with sural nerve problems, as they all spontaneously resolved. Spontaneous biodegradation of resorbable thread and thus probably relieving the sural nerve problems, which might be entrapped in the scar formation in the first weeks in the healing process, what could explain also occurrence of this problem in conservatively treated patients [30, 41, 60], might benefit to the procedures like stretching and friction massage during physiotherapy after healing.
Functional treatment with early ankle motion and early weightbearing could diminish negative effect of immobilization and thus risk for deep venous thrombosis (DVT) and venous thromboembolism (VTE). There is but no clear consensus about incidence of VTE and prophylaxis in patients after ATR [11, 109–111]. The reported overall incidence for VTE in foot and ankle surgery with and without chemoprophylaxis was 0.6% (95% CI 0.4–0.8%) and 1% (95% CI 0.2–0.7%) [109]. Calder et al. found greater risk of VTE in patients with ATR with a clinical incidence of 7% (95% CI 5.5–8.5%) [110]. Aufwerber et al. reported about 35 (37%) of 94 patients DVT in the early functional group with full weightbearing and ankle motion in orthosis and 14 (29%) of 49 patients in control group with 2 weeks of unloading in plaster cast followed by 4 weeks in weightbearing in orthosis, 6 weeks after ATR [111]. Patel with co-workers reported but lower overall rates in patients with ATR for DVT of 0.43% (5 of 1172) and for pulmonary embolism of 0.34% (in 4 of 1172 patients) [109]. The American College of Chest Physicians' (ACCP) most recent review recommends against chemical prophylaxis in lower leg injuries requiring immobilization [112]. It is but necessary that patient-specific risk factors for VTE should be used to assess patients individually [110, 112]. In our series with no routine thromboprophylaxis in any of groups and with no patients on routine anticoagulant therapy, there was no case of DVT and no case of pulmonary embolism. As the operation was performed in local anesthesia with infiltration, patients who might have been on peroral anticoagulants should be switched before operation to low molecular weight heparin (LMWH) and reverted to peroral anticoagulants in the next days after the surgery.
Wu et al. found in a systematic review and network meta-analysis of 2060 patients with ATR in 29 randomized controlled trials the mean incidence of overall major complications from all managements 9.13% (median, 6.67%) and 8.47% in the group of minimally invasive surgery and accelerated rehabilitation [46]. The mean incidence rates of rerupture, deep infection and DVT from all managements were 5%, 1.50%, and 2.67%, respectively [46]. The results in our study, with the use of modified AT repair under local anesthesia and functional treatment, showed lower incidence of any of described complications (6.2% of overall complications, with no re-ruptures, no infections and no DVT).
Despite studies support its efficacy in any type of treatment of ATR [5, 9–11], early functional rehabilitation has lacked a standard definition and interventions and outcome measures are highly variable [113]. There’s also limited evidence for optimized rehabilitation regimen and guidelines, particularly for the first 6 weeks after ATR [113, 114]. As it is associated with a lower complication rate and achieves superior and more rapid functional recovery than conventional immobilization, it was proposed that early ankle motion combined with early weightbearing should become the standard rehabilitation protocol after surgical treatment of acute ATRs [48, 113]. This concept should be therefore accepted also in percutaneous (minimally invasive) treatment. It should be kept in mind that during early functional rehabilitation repaired ATs are exposed to a lot of (cyclic) loading, so gapping and tendon lengthening could occur with too aggressive burdening. A very simple clinical advice could be therefore proposed to patients, to perform the extent of ROM and load during weightbearing until pain is felt. It means but on the other side that patients should be compliant and motivated in rehabilitation process what could be generally very well seen in (professional) sportsmen, who are sometimes even too eager in exercising with even too much loading of the operated tendon, what increases the risk of re-rupture and tendon elongation. Patients in our study in both groups started immediately after the procedure with toe-touch technique of walking with crutches and within the first week with partial weightbearing about 5 kg. Within the first 3 weeks after the procedure they were allowed to increase partial weightbearing up to 15 kg and after 3 weeks to put weight as much as tolerated (until pain). Patients in FG were enhanced to start immediately with ROM exercises as much as their immobilization allowed, which was designed in the way that enabled immediately PF, but restricted DF. Immobilization that was made of stockinette and one package of softcast in FG and 2 packages in IG (to be rigid) was not only very cheap, but also comfortable to wear and simple to take off and on in patients in FG, who were allowed to do this after a week and even to perform PF and DF in water without orthosis until pain was felt. Bilateral seated exercises were allowed in both groups after 3 weeks, when softcast was changed into neutral position of the foot in IG. Putting the weight on the foot produced softening of the dorsal splint in patients in FG and enabled gradual increasing of DF till neutral position, what served also as an effective control of weightbearing in patients in this group. There was no breakage of immobilization in any patient within 6 weeks until it was finally removed in both groups. There was no increase in DF or PF at the end in any of the patient and there were 2 patients in both groups who loss DF for up to 10° and 3 patients in both groups with loss of PF up to 10°. Increased DF and loss of PF could reflect increased length of AT, but all the patients in our series with diminished PF at the final examination had also diminished DF, what could probably reflect more a stiffness of ankle joint and/or capsular adhesions than a tendon elongation. Patients in FG reached earlier correct pattern of gait without limping, earlier better strength and final ROM and were more satisfied with treatment, but the final results according to ROM, strength, return to previous and sports activities and subjective assessment were comparable in both groups with no statistically significant differences.
Many different scoring systems have been proposed for assessment of clinical outcome after ATR treatment but none has been universally accepted [4–6, 26, 79–82, 108, 115, 116]. There are many reasons for that, including subjective parameters in some scales and high technical demands and costs in others [11, 60, 80, 82]. AOFAS hindfoot-ankle score [79] is one of the attempts to solve this dilemma, showing good reliability and validity compared with other scoring systems, despite still including some subjective determinations and parameters less relevant to ATR treatment [80, 81]. Some adapted scores such as the Achilles tendon Total Rupture Score [115] and Self Reported Foot and Ankle Score (SEFAS) [116] had not been developed and reported when our randomized prospective study started, so we used AOFAS hindfoot-ankle score, which has been used in many other studies [22, 26, 33, 117]. The average AOFAS hindfoot-ankle scores of 96.87 in FG and 95.96 in IG (p > 0.005) showed no statistically significant differences between both groups and are comparable to the results of other percutaneous and minimally invasive methods [22, 26, 33, 117].
One of valuable objective parameters in assessment of tendon strength after the treatment is isokinetic testing [5, 82, 118–121]. As this is associated with many, particularly logistic, technical and financial factors, this testing is not universally accepted in assessment scores [5, 6, 30, 82]. Isokinetic studies in other studies showed no statistical difference in strength, power or endurance between open and percutaneous repair [57, 119–121]. As the AT, together with soleus and both gastrocnemius muscles (musculotendinous complex), provides ability to raise on toes [85, 122], heel-rise test was used as the basic idea of clinical testing of the strength and functional outcome after ATR [30, 53, 54, 84]. Todorov et al. found in their study that device independent measures, like ROM and amount of heel raise are an excellent tool providing similar information compared to isokinetic testing and could be used to evaluate clinical outcome after ATR [121]. Lunsford and Perry proposed 25 repetitions for a grade of normal, when using the standing heel-rise test [84]. Silbernagel with co-workers found good validity and greater ability to detect differences between the injured and the uninjured sides with the heel-rise work that measures not only the number of heel-rise repetitions, but also with height of heel-rise repetitions and comparison to the noninjured side (Limb Symmetry Index) [53, 54]. Using these criteria, we found at the final results in our series 74.19% of patients in FG and 73.33% of patients in IG (73.77% altogether), who regained symmetrical strength after modified percutaneous repair of the ruptured AT under local anesthesia, but also 4 (6.55%) of 61 patients who were not able to pass the heel-rise test (Test 2) even with the uninjured leg. So, finally there were only 2 (3.27%) of 61 patients who exhibited reduced strength in a way, that they were not able to pass the heel-rise test with an injured leg, but were able to do this with a non-injured leg.
Beside scoring systems assessment of a patients' functional recovery in comparison to their previous activities could be a simple and useful approach in assessment of different methods, treatment protocols and outcome. Lea and Smith took a very basic approach to outcome; if the patient returned to the preinjury activity level, the outcome was considered good [1]. The results of the presented study were very encouraging in this respect, because 80.64% of patients in FG and 80.00% of patients in IG returned to their previous activities with no limitations and 19.35% of patients in FG and 20.00% of patients in IG reported some difficulties, with no statistically significant differences between both groups.
It should be stressed but that the return of strength, ROM and return to work and previous activities after ATR is very much dependent on the patient's interest and motivation and factors like working contract or litigious and compensation reasons. These factors may influence the final outcome much more than for instance the type of treatment or postoperative rehabilitation. These reasons also speak to why we didn't analyze work absence or time to return to activities as it would be difficult to compare professional athletes or those patients motivated to return to work with patients whose insurance issues allowed them to benefit from a lack of recovery.