Mini-open carpal tunnel release: technique, feasibility and clinical outcome compared to the conventional procedure in a long-term follow-up

We sought to evaluate the findings of our anatomically landmarks based mini-open procedure (MCTR) through a palmar approach and to compare its outcome and practicability to the conventional method (OCTR). The study consisted of 100 matched patients (n = 50 MCTR, n = 50 OCTR) with a minimum follow-up of three years. The outcome was characterized via the Disabilities of Arm, Shoulder and Hand Score (DASH), Symptom Severity Scale (SSS), Functional Status Scale (FSC), and Visual Analogue Scale (VAS). All adverse events were observed. An alpha of 0.05 and a confidence level of 95% were set for statistical analyses. Both techniques showed comparable functional results in a long-term period (mean follow-up MCTR: 60 months and OCTR: 54 months). MCTR versus OCTR at mean: DASH: 4.6/8.3 (p = 0.398), SSS: 1.3/1.2 (p = 0.534), FSC: 1.3/1.2 (p = 0.617), VAS: 0.4/0.7 (p = 0.246). The MCTR convinced through a lower rate of scar sensibility (MCTR: 0% vs. OCTR: 12%, 0/50 vs. 6/50; p = 0.007) and pillar pain, as well as a shortened recovery period and surgical time relative to the OCTR. Low complication rates were observed in both groups, no recurrences had to be documented. The MCTR procedure revealed a similar good clinical outcome as the conventional technique. MCTR is a minimally-invasive, reliable, fast and simple procedure with an obvious benefit regarding scar sensibility.

The aim of this study was to evaluate the long-term follow-up of a MCTR based on previously described anatomically landmarks 14 via a palmar approach, and to compare it to the conventional OCTR. Therefore, we hypothesized that the safety and practicability between MCTR and OCTR is similar. More specifically, we sought to determine if MCTR is superior to traditional techniques in terms of patient satisfaction and functional recovery.

Material and methods
Patient collective. Fifty patients, who had undergone unilateral MCTR during the time interval between 1st January 2008 and 31st December 2015 at our Level III trauma center, were included in the study. Fifty patients, who had received conventional OCTR during the same time stage, were matched regarding age and sex to the MCTR group. Each surgical group had been treated by one experienced trauma consultant and hand specialist (MCTR group: MP, OCTR group: AK). The type of operative procedure was not randomized. A minimum follow-up period between surgery and patient visit of three years was defined for final reevaluation.
The diagnosis of CTS was founded on the occurrence of sensory disturbances and/or weakness alongside the supply territory of the median nerve as well as a distinctive anamnesis of pain. All patients obtained a pre-operative electromyogram testifying median nerve neuropathy. Conservative treatment including previous physiotherapy and splint application failed in all patients.
Exclusion criteria involved local or systematic signs of inflammation, distortion of anatomy, neurologic or soft tissue defects, former wrist and hand surgeries as well as bilateral CTS symptomatic and durance of symptoms of more than one year. Patients with other co-morbidities like insulin dependent diabetes mellitus, polyneuropathies, smoking or rheumatoid arthritis were omitted.

Surgical techniques.
A standardized scheme was used for both groups. Surgery was performed in a consistent technique per group, any specific differences per surgical procedure are described below. All surgeries were performed in the operation theatre in regional anesthesia and by using an upper arm tourniquet. All hands were placed in supine position and fixed with a lead hand splint. A mini-suction drainage was inserted and systematically left in place for one day. Wound-closure was performed using non-absorbable 5.0 sutures {Polyamide, ETHILON 668H, Johnson & Johnson Medical GmbH, Norderstedt, Germany} without the use of Leukostrips. All wounds were covered with a Gauze Swab {Lohmann & Rauscher International GmbH & Co. KG, Rengsdorf, Germany} for direct compression. Before releasing the upper arm tourniquet, a dressing was applied consisting of Gauze Swabs as a padding roll following a medium-stretch Compression Bandage {Lenkelast, Lohmann & Rauscher International GmbH & Co. KG, Rengsdorf, Germany}. After the surgical procedure, the hands were kept in this fist bandage for one day. Early dressing removal, including mini-suction drainage removal and clinical control were conducted via the surgeon 24 h postoperatively. Then, a plaster {Cosmopor, Hartmann AG, Heidenheim, Germany} was applied and the patients were discharged. The free-functional aftercare was started on the first postoperative day, avoiding heavy weight lifting for two weeks. The sutures were removed 14 days following surgery (first follow up).
OCTR . Concerning the OCTR technique, a longitudinal incision of 3.5 cm was performed alongside the median palmar crease in a proximal direction and stopped 0.5 cm distally to the wrist crease. Following subcutaneous dissection, the transverse carpal ligament, as well as the thenar branch of the median nerve, were depicted and the ligament was cut at its ulnar border in a proximal direction.

MCTR .
Regarding the MCTR group, both styloid processes were palpated and connected with a horizontal line by use of a pen. Then, a parallel line was drawn in the palm from the most proximal extend of the first web space, named Kaplan's cardinal line. This line was addressed as a surface marker for a safe zone to the superficial palmar arch in the literature previously [15][16][17] .
Next, a perpendicular longitudinal line was drawn on the ring finger's radial side. This line ends directly over the tendon of the palmaris longus muscle on the distal forearm, if existing. The intersection point of these lines in the palm was used as reference point (A). The skin incision extends the one-third from the reference point distally, the other two-thirds in a proximal direction (see Figs. 1

and 2).
A 1-1.5 cm palmar incision was performed in a longitudinal fashion proximally, passing in every case the thenar crease from ulnar. Following the opening of the palmar aponeurosis, the transverse carpal ligament was tunneled above and below itself in a proximal direction with Hegar dilatators (see Fig. 3). The superficial palmar arterial arch as well as the thenar branch of the median nerve were depicted and protected (see Fig. 4).
The first incision of the distal portion of the transverse carpal ligament was performed under direct vision with a surgical scalpel blade number 15 {Dahlhausen & Co. GmbH, Köln, Germany}. Next, the protective guide was inserted into the carpal tunnel in a proximal direction underneath the remaining transverse carpal ligament. The special cutting knife was inserted into its groove and passed proximally until a complete release of the remaining transverse carpal ligament has been accomplished (see Figs. 3, 5 and 6).

Reevaluation.
Retrospectively evaluated data were collected prospectively in a dedicated hospital database.
Basic characteristics, which involved age, sex, affected side, duration of surgery and pre-surgical symptoms were recorded from the respective patient files.
Patients had a follow-up visit at two weeks following surgery. The early follow-up was assessed by the respective surgeon (MP or AK). Here, the presence of wound healing disturbances, infection, scar sensibility and the respective Visual Analogue Scale (VAS) were evaluated.
The final follow-up was done by two trauma surgeons (AS and GH), documenting changes in symptoms and signs and/or adverse events. During the final follow-up, all patients completed the VAS, the Disabilities of Arm, . Additionally, the patients were questioned to the occurrence of scar sensibility and/or pillar pain at the site of the operation. The evaluation of scar sensibility was defined as follows: The patients were questioned about complaints at the area of the scar like burning discomfort, hypersensitivity or superficial pain during palpation. Scar sensibility was classified as the presence of one or more symptoms. Pillar pain was defined as deep-seated pain of the thenar and/or hypothenar region, related to the use of the hand like tight gripping. During examination, pillar pain was tested by simultaneous compression of thenar and hypothenar eminences as if to separate the carpal tunnel. The durance from surgery to restitution and return to workplace (retirees excluded) was assessed. Restitution was defined as the subjective time following surgery when patients were able to perform their activities of daily living painlessly. The occurrence of adverse events (hematoma, infect, neurovascular or tendon lesions) and recurrence were evaluated. Recurrence was constituted by the return of symptoms after a temporary period of resolution and/or the need for a symptom-associated reoperation. Further, the revision rate and cause as well as the time interval from primary surgery to the revision surgery, if applicable, were assessed. Statistical analysis. Statistical analyses were performed using the SPSS software {IBM SPSS Statistics version 26, Armonk, USA}. Continuous parameters were presented as mean, standard deviation (SD), and categorical or quantitative data. Descriptive statistics were used for demographic variables; the continuous variables were summarized using SD and/or range via minimum and maximum.
Non-parametric tests were used for data analysis regarding significance. To investigate the differences between the MCTR and OCTR group, Mann-Whitney U tests were utilized. Chi-square tests were used to compare the  Ethical details. Ethical approval was obtained from the institutional review board of the Austrian Workers' Compensation Board (AUVA-EK 03/2019). While included, the patients consented to the study protocol and informed consent was obtained for research purposes. All the experimental protocols and methods were carried out in accordance with the regulations and principles of the Declaration of Helsinki and the ICH-GCP Guidelines. There were statistically significant differences between the durance of surgery (p = 0.001), but no statistically significant differences between the groups regarding age (p = 0.621), and time of follow-up (p = 0.623).

Final follow-up (clinical results and adverse events).
None of the evaluated scores differed statistically significantly between the groups (see Table 1). All postoperative progresses can be overlooked in Table 2. Of these, scar sensibility was significantly decreased (p = 0.007) in the MCTR group (0%) when compared to the OCTR group (12%, 6/50).
The perioperative adverse events are listed in Table 3. In both groups, no recurrences were to observe. The revision rate was 2% in the MCTR group (1/50) and 4% in the OCTR group (2/50). In both samples, patients suffering from an infection underwent one singular revision surgery. The mean time from primary to revision surgery was two weeks for the MCTR and 6.3 weeks (range: 3.5-9) for the OCTR group.
In the overall collective, no iatrogenic vascular, nerve branch or tendon injuries were documented. One partial median nerve lesion on the palmar aspect was to verify in the MCTR group, following by extending the incision and direct nerve repair via micro-neurosurgical technique. A full nerve function could be reached two www.nature.com/scientificreports/ years postoperatively, which was evaluated clinically and via electromyogram testifying. In total, the overall complication rate amounted to 4% (4/100 in total or 2/50 per group).

Discussion
The aim of this study was to evaluate the long-term follow-up of a minimally-invasive CTR through a palmar approach and to compare its outcomes to the conventional procedure. In our sample, none of the functional scores differed significantly between the groups. The scar sensibility was statistically significantly decreased following MCTR when compared to OCTR in the early-(p = 0.002) and long-term follow-up (p = 0.007). Paine 20 has descript the first device used as a retinaculotome in 1955. This device has been used up to now, as Fernandes et al. 21 reported more recently. The authors 21 observe short-and long-term results and reported good clinical results of more than 500 patients in a period of 17 years. The palmar incision for CTR has been used for a long time, various strategies have been reported over the time. Aryan et al. 22 reported 1977 a representative case series of 429 patients with improved symptoms. Using a similar retinaculotome, our findings address a lack in the literature. Existing studies 23,24 are scarce regarding long-term outcomes following MCTR.
Bai et al. 2 performed a retrospective analysis of prospectively collected data concerning 85 patients who had either undergone MCTR or OCTR including similar incisions as performed in the current study. The average duration of symptoms had been 6.6 months (MCTR) and 6.4 months (OCTR), respectively, which is comparable to our sample (MCTR: 4.9 months; OCTR: 5.4 months). Mean durance of surgery did not differ significantly (p = 0.130) between the groups (MCTR: 25.1 min; OCTR: 23.5 min), which was different in our collective (MCTR: 9.2 min; OCTR: 12.0 min; p = 0.001). As in our sample, the authors found no statistically significant differences     25 , divided their sample of 105 patients into three subgroups (MCTR, OCTR and a group undergoing endoscopic CTR). The average interval of return to work was significantly (p = < 0.05) longer in the OCTR (mean: 21.1 days) when compared to the MCTR (mean: 12.7 days). We also evaluated a shorter durance of absence at workplace and a non-statistical significant superiority of the MCTR group (MCTR: mean 14 days; OCTR: mean 20 days; p = 0.142). This could be considered as positively cost-effective on the basis of an earlier return to work.       23 performed an analysis of 207 patients who were randomized into a MCTR group through two small incisions (n = 73), an OCTR group (n = 65) and a group that underwent endoscopic CTR (n = 69). The mean duration of symptoms had been six months in the MCTR and OCTR group, which is well comparable to our sample (MCTR: 4.9 months; OCTR: 5.4 months). At the final follow-up of three years, there were no statistically significantly differences between MCTR and OCTR regarding outcomes of the Boston Carpal Tunnel Syndrome Questionnaire. Here, the SSS and FSC were at a mean of 1.2 points in both groups. These values are well comparable to our results.

Scar sensibility (in cases) Pillar pain (in cases) Restitution (in weeks) Return to work (in days)
Recurred nerve compression occurs at a rate of less than 2% to as high as 25% [26][27][28][29][30][31] and may happen years after surgery 30 . Recurrence rates in long-term follow-up studies are reported from 3.7% 27 to 57% 32 of cases. However, there is a lack of consistent definition for recurrence in the literature, which may explain this range. Some authors define the return of any preoperative symptoms 32 , others the need for reoperation 27 to qualify as recurrence. Cresswell et al. 24 reported a higher rate of immediate complications, and more recurrences in patients following MCTR compared to the conventional procedure. The authors 24 assessed results seven years postoperatively. In contrast, no recurrences were to observe in both of our groups with a minimum follow-up of three years and a shorter final follow-up (mean final follow-up of MCTR: 60 months and OCTR: 54 months). This might be traced back to our shorter follow-up intervals and the bias on the returned questionnaires of Cresswell et al. 24 , which may not reflect the entire cohort.
Common and main indications of recurrences of nerve compression may cause the incomplete splitting of the transverse ligament and/or a postoperative fibrosis 30,33 . It is to believe, that the postoperative fibrosis may positively influenced, like with our early motion protocol from the first postoperative day on. Complete division of the carpal ligament is a factor associated with surgery. Kilinc 34 showed, that recurrent CTS after sufficient division of the transverse ligament is very unlikely. Additionally, similar low adverse events were observed in both groups. Anyhow, these facts indicate the positive effectiveness and permit the inference, that MCTR represents a reliable method for CTR.
One further technical advantage represents the direct visualization of the thenar branch of the median nerve. Further, no iatrogenic lesions of the superficial palmar arterial arch can occur due to the proximally directed release. In this regard, we highlight, that we had to observe one partial median nerve laceration in the MCTR group in one patient with high-grade adhesions around the median nerve. Similarly, Cresswell et al. 24 reported one lesion of the median nerve in 53 patients. Lee and Strickland 17 observed two median nerve lesions in their 694 releases with the retinaculotome. Summing up our findings and both studies, the median nerve lesion is a noteworthy major complication, whereby the rate represents less than 1% in all three research groups. Based on our observations and experiences, we strictly recommend to enlarging the skin incision in patients with highgrade adhesion situations.
This study had several limitations. We solely compared MCTR with OCTR and did not include an endoscopic group. Moreover, no randomized assignment was calculated for the respective groups. Furthermore, although we used prospectively collected data, the study remains retrospective nature. Randomized controlled trials would be recommendable to re-evaluate this technique in future. Minimally-invasive procedures are generally preferred by patients, although patients as well as hand surgeons may have decisional conflicts about the treatment 35 . These factors were not addressed in the study. Lastly, financial implications were not reviewed regarding material costs or the localization of doing the surgery or inpatient versus outpatient setting.
As strength of this study is the long-term timeframe to entitle. To substantiate the operative technique in special, we used a matched-patient as well as single-surgeon study design with experienced practitioners. Moreover, the surgical procedure is easily reproducible, because it is based on superficial anatomical landmarks. The further magnitude of this minimally-invasive method represents the simple approach and the intraoperative visibility of both structures at danger-namely the thenar branch of the median nerve and the superficial palmar arterial arch. Our method was developed under inclusion of well-known anatomical safe zones [15][16][17] and anatomical landmarks for a mini-open technique as proven by Hohenberger et al. in an anatomical study 14 .
In conclusion, our suggested and preferred MCTR technique through a palmar approach has been demonstrated to be effective. Patients following MCTR reached the same functional long-term outcome as the conventional procedure. The MCTR procedure has a low complication rate, both techniques are comparable in this respect. It is to recommend to enlarging the skin incision in patients with high-grade adhesions.
With the technique we described, no recurrences or patients suffering from pillar pain were to observe. Additionally, it combines the advantages of a reduced recovery period and surgical time. The major patientspecific benefit represented the decreased scar sensibility due the small 1-1.5 cm incision and its localization, which convinced as statistically significant when compared to the OCTR group. These positive effects might be traced back to the reduced damage of soft tissues while preserving the adjacent structures attributed. Thus, MCTR is a fast, practicable, minimally-invasive and minor technically challenging procedure, which allows a direct visualization of anatomical structures at risk.