Endoscopic Carpal Tunnel Release

Techniques, Controversies, and Comparison to Open Techniques

Jacques H. Hacquebord, MD; Jeffrey S. Chen, MD; Michael E. Rettig, MD


J Am Acad Orthop Surg. 2022;30(7):292-301. 

In This Article

Endoscopic Versus Open Carpal Tunnel Release

Symptom Relief

The Boston Carpal Tunnel Syndrome Questionnaire (BCTQ) is a validated, two-part patient-reported survey divided into symptom severity (BCTQ-S) and functional status (BCTQ-F) portions, which is frequently used to assess symptom severity in carpal tunnel syndrome.[16] A meta-analysis of seven studies using the BCTQ to compare ECTR and OCTR found no differences at 1 year with either BCTQ-S (mean difference [MD] 0.15; 95% CI [−0.04, 0.35]; n = 592; P = 0.13) or BCTQ-F (MD 0.17; 95% CI [−0.02, 0.36]; n = 592; P = 0.08).[17] Similarly, there were no differences in pain relief between ECTR and OCTR as measured by the Visual Analogue Scale before or after 3 months postoperatively.[18]

No significant differences were found between ECTR and OCTR with respect to digital sensation with Semmes-Weinstein monofilament testing at 3 months (MD 0.06; 95% CI [−0.09, 0.21]; n = 297; P = 0.43) or with two-point discrimination testing at 1 year (MD −0.16; 95% CI [−0.45, 0.12]; n = 402; P = 0.26).[17] Other studies using Semmes-Weinstein monofilament testing as a surrogate for paresthesia relief had similar findings with no differences when further stratified into less than 3 months and greater than 3 months.[18]

Sleep disturbance is a common manifestation of carpal tunnel syndrome. To study sleep disturbance specifically, Gaspar et al randomized 60 patients to ECTR or OCTR and administered the Pittsburgh Sleep Quality Index (PQSI), the Insomnia Severity Scale, and the Quick Disabilities of the Arm, Shoulder and Hand surveys at three postoperative time points. They found that ECTR provided superior improvement in the PQSI (P = 0.016) and Insomnia Severity Scale (P = 0.006) at the first postoperative visit (1 to 2 weeks) and PQSI (P = 0.0038) at the second postoperative visit (4 to 6 weeks), with no differences at final follow-up (6 to 12 months).[19] Thus, although both ECTR and OCTR can alleviate sleep-related symptoms, ECTR may do so more rapidly.


A prospective randomized trial by Trumble et al[20] including 192 hands found that patients treated by ECTR had significantly greater grip strength, pinch strength, and hand dexterity during the first 3 months after surgery (P < 0.05). One meta-analysis similarly concluded that ECTR results in significantly better pinch strength than OCTR at 3 months postoperatively.[17] Another meta-analysis found similar results initially; however, once the studies were filtered and only long-term data (defined as greater than 6 months postoperatively) were included, there were no differences between ECTR and OCTR with either grip strength (MD 0.90 kg; 95% CI [−1.47, 3.27]; n = 571; P = 0.46) or pinch strength (MD 0.37 kg; 95% CI [−0.09, 0.84]; n = 614; P = 0.12) at long-term follow-up.[21] These results paralleled those of Trumble et al,[20] who found that the better strength results initially seen with ECTR tapered off and were equivalent to those of OCTR after 3 months. A possible explanation is that smaller skin incisions and less soft-tissue dissection with ECTR result in quicker soft-tissue recovery and earlier confidence to resume activity and begin strengthening.

Return to Work

In line with the reasoning that less soft-tissue violation may lead to earlier improvements in strength and dexterity, ECTR has been theorized to result in quicker return to work times, one of its more impactful proposed advantages. Meta-analyses agree that ECTR results in significantly shorter return to work, the largest of which reviewed 11 studies and found a mean difference of −8.73 days (MD −8.73 days; 95% CI [−12.82, −4.65]; n = 1,234; P < 0.001).[21] Other studies have shown mean differences ranging from −7.25 to −9.56 days.[17,18,22] However, these pooled analyses did not make distinctions regarding patient occupation, which is an important distinction as surgeons may have different return to work criteria for manual laborers than for those working desk-based jobs. Accordingly, the authors found a large degree of heterogeneity between studies, which may be attributed to this variability of work and daily activity.[21,22] Despite these limitations, there does seem to be a significant difference in return to work between these techniques, which has broad implications when considering cost-effectiveness of the two procedures. Future studies stratifying results by type of work and functional demand may lead to a deeper understanding of this finding.


Complications as a whole are difficult to study as they are relatively rare in nature and depend heavily on surgeon reporting, patient follow-up, and even the definition of a complication itself. Early studies on ECTR reported unacceptable complication rates ranging from 2% to as high as 35%.[23] With further experience and advancements in endoscopic technique, more recent studies have reported complication rates similar to those of OCTR, although controversy remains.[24]

Benson et al performed a literature review of 68 articles spanning over 27,000 cases of carpal tunnel release to study rates of complications, which they defined as damage to nerves, arteries, or tendons. The authors found a higher overall complication rate with ECTR compared with OCTR (1.63% vs 0.74%, P < 0.005). On further analysis, a large proportion of complications associated with ECTR were transient neurapraxias, which self-resolved. In fact, ECTR was associated with a 1.45% rate of transient neurapraxia versus 0.25% with OCTR (P < 0.005). When transient neurapraxias are removed from the analysis, the overall rate of structural complications is actually significantly lower with ECTR compared with OCTR (0.19% vs 0.49%, P < 0.005).[25]

Vasiliadis et al performed a meta-analysis explicitly focused on safety and complication risk, stratified into major and minor complications. Major complications included permanent structural injury, complex regional pain syndrome, and moderate or severe pain at greater than 1 year follow-up. Minor complications included any postoperative issues, which resolved at final follow-up (ie, scar issues, pillar pain, and transient neurapraxia) and delayed recovery continuing up to the latest follow-up reported. The authors reported no differences between ECTR and OCTR with respect to major complications (odds ratio [OR] 1.00; 95% CI [0.44, 2.27]; n = 2,565; P > 0.05), with a very low overall major complication rate (0.9%). On the other hand, ECTR was associated with 50% decrease in odds of minor complications compared with OCTR (OR 0.50; 95% CI [0.31, 0.82]; n = 2,442; P < 0.05). Specifically, ECTR carried a 76% decrease in odds for wound or scar issues compared with OCTR (OR 0.24; 95% CI [0.15, 0.40]; n = 1,943; P < 0.05) but 2.42 times the odds for transient neurapraxia (OR 2.42; 95% CI [1.22, 4.80]; n = 2,182; P < 0.05).[22] These results were mirrored in several other meta-analyses.[17,18,21]

One hypothesis for the increased rates of transient neurapraxia with ECTR is the brief increase in tunnel pressure, and thus pressure exerted on the median nerve, associated with insertion of instrumentation before ligament release. Uchiyama et al used an intraoperative pressure transducer while performing two-portal ECTR and found that tunnel pressure was greatest immediately before the cannula was withdrawn from the exit portal. However, no conclusion can be made regarding the relationship of tunnel pressure to postoperative neurapraxia as the transducer was only accurate to 300 mmHg (nearly all maximum pressures were >300 mmHg) and no cases of postoperative neurapraxia were reported.[26] In addition, single-portal and two-portal techniques may be subjected to different pressure environments as the wrist is suspended in hyperextension in the latter. Additional studies are necessary to determine whether this is truly the etiology of transient neurapraxias. Increased rates of wound and scar-related issues with OCTR can be explained by the larger surgical incisions made directly over the base of the volar palm.

Incomplete release of the TCL is sometimes defined as a complication of carpal tunnel release and is an inherent concern of ECTR due to potential limitations with endoscopic visualization. Incomplete release is difficult to measure directly. Failure to relieve symptoms due to persistent compression is a proxy measurement, which was previously explored in this discussion and shown to have no differences between the two techniques. Rate of revision surgery is another proxy measurement, although less reliable as revision surgery can be performed for a variety of reasons not limited to incomplete release. Chen et al[18] analyzed seven trials and found no significant increase in risk of revision surgery following either procedure (relative risk 1.18; 95% CI [0.53, 2.63]; n = 1,031; P = 0.68). Other meta-analyses also found no significant increase in risk or difference in incidence of revision surgery.[21,22]

The methodology through which complications are assessed must be considered. As previously mentioned, a large proportion of data are based on surgeon reporting. Take for example, a study conducted by Devana et al using the PearlDiver database to investigate the billing records of over 60,000 patients. The authors found a lower rate of median nerve injury after ECTR versus OCTR in both private payer (0.59 vs 1.69 per 1,000 patients) and Medicare (1.96 vs 3.72 per 1,000 patients) groups.[1] However, these conclusions were made based on International Classification of Diseases, Ninth Revision codes and are heavily dependent on proper and appropriate coding.

On the other hand, Trehan et al used a New York State database to investigate rate of subsequent nerve repair specifically, rather than all-cause revision surgeries, by searching for Common Procedural Terminology codes for nerve repair procedures following carpal tunnel release. Their study had the benefit of a state-wide registry, which was able to capture data by patient rather than by provider, including patients who received revision surgery by another provider or at another institution. The authors found that although low overall, the rate of subsequent nerve repair after ECTR was significantly higher than OCTR (0.09% vs 0.04%, P = 0.01) and carried a hazard ratio of 2.40 (HR 2.40, 95% CI [1.02, 5.27]).[27] Their study was limited by state boundaries and accurate procedural coding, but nonetheless introduces the idea that the true incidence of nerve injury may not be so simple to measure.

Direct Cost

On the surface, it seems obvious that ECTR would be more expensive than OCTR. Koehler et al[28] used a time-driven activity-based costing technique to characterize costs and found that total procedural costs of ECTR were 43.9% greater than OCTR ($2,759.70 vs $1,918.06), largely related to disposables, direct operating room costs for procedural duration, and physician labor. Other studies using different accounting methods all generally agree that the immediate procedural costs of ECTR exceed those of OCTR, some estimates by as much as $2,000 per case.[29,30]

Direct costs are not limited to the choice of procedure, but also encompass surgical setting (procedure room versus operating room) and anesthesia choice (local, monitored anesthesia care, Bier block, and general). Kazmers et al used their institutional data to calculate total direct costs of carpal tunnel release surgical encounters and found that performing OCTR in the procedure room under local anesthesia was the least costly technique. Performing OCTR in the operating room under monitored anesthesia care was approximately 11 times as costly, whereas ECTR in the same setting was more than 15 times as costly.[31] Certainly, the exact surgical environment is limited by the inability of some patients to tolerate local anesthesia alone or by the availability of equipment and anesthesia options in some settings, but these factors must all be considered together when discussing direct costs.


The differences are less clear, however, when looking at cost-effectiveness rather than direct upfront cost. To fully understand the idea of cost-effectiveness, it is important to realize the societal cost of carpal tunnel syndrome and its surgical treatment. Data reported by the US Bureau of Labor Statistics show that carpal tunnel syndrome leads to a median of 30 days away from work.[32] A 5-year analysis of workers' compensation trends in Washington State estimated $310 million in claims related to carpal tunnel syndrome, of which $149 million was allocated to temporary disability.[33] Thus, when considering cost-effectiveness, the cost to society from loss of productivity and opportunity cost from missed work must be considered.

Barnes et al used a Markov model to perform the most recent comprehensive cost-effectiveness analysis of ECTR versus OCTR. With the assumption that patients undergoing ECTR were able to return to work 8.21 days earlier than those undergoing OCTR, they found that ECTR was actually more cost-effective than OCTR from the societal perspective.[34] ECTR was also more cost-effective from the payor perspective when using a willingness-to-pay threshold of $100,000/quality-adjusted life year, a common threshold used in healthcare cost-effectiveness studies.[34,35] Specifically, ECTR in the office setting is the most cost-effective measure, followed by ECTR in the operating room, OCTR in the office setting, and OCTR in the operating room. Cost-effectiveness studies have many inherent limitations including numerous variables and assumptions gathered from other studies, but are still frequently used to justify reimbursement of procedures. As the United States continues to transition toward value-based delivery of health care, the cost-effectiveness of procedures performed will undergo continued scrutiny.


One last cost-related factor to consider is physician compensation. Currently, distinct Common Procedural Terminology codes exist for both OCTR (64721—neuroplasty and/or transposition; median nerve at carpal tunnel) and ECTR (29848—endoscopy, wrist, surgical, with release of transverse carpal ligament), each with its own assigned work relative value units (wRVU). OCTR is currently valued at 4.97 wRVU, whereas ECTR is currently valued at 6.39 wRVU, a 29% increase.[27] Along these lines, a surgeon's salary structure (ie, incorporating wRVU or not) and any ownership or stake in facilities may sway individual decision making. This should be kept in mind when interpreting studies or opinions of others.

Patient Satisfaction

A handful of studies have reported on patient satisfaction following ECTR versus OCTR. As there are virtually no validated tools for measuring patient satisfaction, these results can be subjective and should be interpreted accordingly.[21] Using a scale of 0 to 100 points with a higher score corresponding to increased satisfaction, analysis of these studies has shown that patients tend to be more satisfied with ECTR than OCTR (MD 3.13; 95% CI [1.43, 4.82]; n = 303; P = 0.0003).[17] Michelotti et al followed a cohort of patients who underwent staged bilateral carpal tunnel release randomized to either ECTR or OCTR on the first side followed by the other technique on the contralateral side. They found no statistically significant differences in patient satisfaction with either procedure.[36]

Surgical Time

Studies are divided as to the surgical time of ECTR versus OCTR. Studies have been published showing that ECTR is significantly faster than, equivalent to, and significantly slower than OCTR.[17,21,28] In reality, surgical time is highly variable and difficult to measure due to discrepancies in location of procedure, type of anesthesia, familiarity of surgical staff with setup, presence of trainees, surgeon skill and experience level, etc. The exact difference in procedural time is likely irrelevant and would not affect cost-effectiveness studies in a significant manner.[34]

Opioid use

Opioid use after carpal tunnel release has also been studied. A retrospective review analyzed trends in postoperative analgesic use and found that patients who underwent OCTR were significantly more likely to fill opioid prescriptions (62% vs 60%, P = 0.034), more likely to refill earlier (16% vs 14%, P = 0.008), and filled higher quantities (411 ± 96 vs 379 ± 82 oral morphine equivalents, P < 0.001).[37] Although statistically significant, these differences were small and their applicability is unclear as other studies have shown no differences in pain scores, coping skills, or symptoms with opioid versus nonopioid analgesic postoperative regimens after carpal tunnel release.[38] Postoperative analgesic use after carpal tunnel release is likely correlated more with individual prescriber patterns rather than chosen technique. A summary of results from all meta-analyses comparing ECTR with OCTR, which included 15 or more randomized trials from 2014 to present, is presented in Table 1.