Abstract
Background: Ultrasound can provide evaluation of the anatomy of the carpal tunnel in a convenient, noninvasive office setting. This study is intended to determine the accuracy and diagnostic performance of ultrasound, used by surgeons, for the evaluation of completeness of carpal tunnel release (CTR). Methods: Ten cadaver arms underwent randomized sectioning of 0%, 25%, 50%, 75%, or 100% of the transverse carpal ligament. Following a brief training session, a blinded observer used ultrasound to evaluate the percentage of the transverse carpal ligament release. The release amount was then confirmed with an open exposure of the transverse carpal ligament. Results: Cronbach α and Pearson correlation coefficients were 0.92 and 0.87, demonstrating excellent reliability and validity of the technique. Diagnostic performance including sensitivity, specificity, positive predictive value, and negative predictive value was 100%, 75%, 86%, and 100%, respectively, for the diagnosis of incomplete release of the transverse carpal ligament by a novice sonographer orthopedic surgeon. Conclusions: The ultrasound is a highly accurate tool for the diagnosis of incomplete transverse carpal ligament release and requires a minimal amount of training to use for this purpose. It provides a rapid means of diagnosing incomplete release of the transverse carpal ligament following CTR.
Keywords: carpal tunnel release, ultrasound, persistent carpal tunnel syndrome, incomplete release
Introduction
Long-term results of open and endoscopic surgical treatment of carpal tunnel syndrome (CTS) have been comparable in multiple studies.1-3 Incomplete release of the transverse carpal ligament (TCL) is the most common reason for persistent symptoms after carpal tunnel release (CTR), requiring revision surgery. In one case series, incomplete TCL release was noted intraoperatively in 50% of revision CTRs.4 In a 1994 cadaver study of surgeons newly introduced to endoscopic CTR (eCTR), 38% of cadavers demonstrated incomplete TCL release. Clinical failure of eCTR, as defined by persistent symptoms, has been reported at rates of up to 6%,5 with the success rate approaching 100% after revision of eCTR with open CTR.6,7
Ultrasonography (US) can effectively visualize the median nerve and the TCL both pre- and postoperatively.8 The US also reliably demonstrates changes in median nerve shape in CTS.9 When performed by a trained sonographer, it has been established as an effective method for the detection of incomplete CTR in symptomatic patients.10
The accuracy of ultrasound is operator-dependent; hence, its use in the diagnostic workup has traditionally been restricted to radiologists and ultrasound technologists. However, ultrasound machines are readily available and easy to use and have a proven ability to observe median nerve and TCL anatomy.8 Thus, we believe that this technique is of use in the hands of orthopedic surgeons in some highly specific applications. In this study, we tested the hypothesis that an orthopedic surgeon, with a basic training in US, is capable of detecting an incomplete CTR. Therefore, we evaluated the diagnostic performance, reliability, and validity of ultrasound performance by novice sonographer surgeons for the diagnosis of incomplete CTR. Confirmation of this hypothesis provides support for use of bedside ultrasound as an initial screening modality for failed CTR.
Methods
All participants in the study (1 senior resident, 1 fellowship-trained hand surgeon, and 2 general orthopedic surgeons) underwent a brief orientation to ultrasound of the carpal tunnel by a certified US instructor. A model of incomplete CTR was created using 10 fresh-frozen cadaver forearms. The demographics of the specimens are shown in Table 1. There was no preexisting evidence of surgical CTR in any cadaver. Each forearm was prepared by the senior author without the involvement of evaluators. A 1-cm transverse incision about 0.5 cm proximal to the volar wrist crease and in tangent to the level of the hamate at the wrist level was made sharply. A distally based flap was created in the flexor retinaculum. Using an eCTR system (S.E.G.-WAY Orthopaedics, Carlsbad, California), the TCL was approached. Under endoscopic visualization and measurement using the endoscopic canula markings, 0%, 25%, 50%, 75%, or 100% of the TCL was divided starting from the distal to the proximal. The amount of ligament released was determined prior to the procedure, and the releases were performed in a randomized order, such that the participants in the study were unaware of the order and blinded to the release performed in each specimen. After removal of the endoscopic equipment, air was manually expelled from the carpal tunnel.
Table 1.
Demographics of Cadaver Subjects.
| Subject | Sex | Age | Side |
|---|---|---|---|
| 1 | F | 78 | R |
| 2 | M | 81 | R |
| 3 | M | 75 | L |
| 4 | F | 38 | L |
| 5 | M | 53 | L |
| 6 | M | 61 | L |
| 7 | F | 52 | R |
| 8 | F | 63 | L |
| 9 | F | 45 | L |
| 10 | M | 60 | R |
Each cadaver forearm then underwent ultrasound with a Mindray M7 portable ultrasound using a 7.5-MHz linear probe (Mindray Medical International Limited, Shenzhen, China) by 2 of the 3 blinded participants, who were asked to evaluate whether 0%, 25%, 50%, 75%, or 100% of the TCL had been transected. In the event of a disagreement, a third observer was used as a tiebreaker (Figure 1). The ultrasound technique consisted of an initial longitudinal view in line with the median nerve, as seen in Figure 1c. The probe was then rotated 90° and slowly advanced from proximal to distal along the course of the carpal tunnel. In the event of an incompletely released carpal tunnel, the nerve was noted to make an abrupt transition between a superficial and a deep position, and a transverse hyperechoic line representing the TCL was noted superficial to it as seen in Figure 1a. The wrist was maintained in neutral position for this entire procedure.
Figure 1.
Ultrasound images of posttransection carpal tunnel. (a) Transverse view of the carpal tunnel following transection of the distal transverse carpal ligament. This image is taken through a transected section of the ligament. (b) Transverse view of the carpal tunnel following transection of the full length of the transverse carpal ligament. (c) Longitudinal view of the carpal tunnel following transection of the transverse carpal ligament.
Note. 1: median nerve; 2: transverse carpal ligament; 3: flexor tendons; line: area of transected ligament.
Following ultrasound analysis, all cadaver samples were subjected to an open carpal tunnel approach. This approach was carried out until the entire transverse carpal ligament or its remnants could be directly visualized. For each sample, the percentage of the TCL that had been transected was measured (Table 2).
Table 2.
Results of the Ultrasound in the Diagnosis of Partial Versus Nonpartial Carpal Tunnel Release.
| Subject number (prior to randomization) | Initial surgical release (% released) | Ultrasound evaluation (% released) | Open surgical evaluation (%) |
|---|---|---|---|
| 1 | 0 | 0 | 0 |
| 2 | 0 | 0 | 0 |
| 3 | 25 | 25 | 25 |
| 4 | 25 | 50 | 25 |
| 5 | 50 | 50 | 50 |
| 6 | 50 | 50 | 50 |
| 7 | 75 | 75 | 75 |
| 8 | 75 | 75 | 75 |
| 9 | 100 | 50 | 100 |
| 10 | 100 | 100 | 100 |
Statistical Analysis
We measured the diagnostic performance of the ultrasound, including sensitivity, specificity, positive predictive value, and negative predictive value (NPV), in the diagnosis of partial versus nonpartial CTR. We pooled 25%, 50%, and 75% release as partial and pooled 0% and 100% release as nonpartial. Subsequently, the results of the ultrasound were considered positive or negative if there was an absolute agreement with the amount of release after open surgical evaluation.
We measured the validity of the ultrasound results using Pearson correlation (correlation between US and open surgical evaluation) to evaluate the diagnostic performance of the ultrasound. We also measured the reliability of the ultrasound results using Cronbach α to check whether there is an internal consistency between the percentage of endoscopic release and the percentage diagnosed by ultrasound. Cronbach α coefficient >0.9 indicates excellent internal consistency; 0.9 > α > 0.8, good; 0.8 > α > 0.7, acceptable; 0.7 > α > 0.6, questionable; 0.6 > α > 0.5; and >0.5, unacceptable.
Results
Diagnostic performance results of the ultrasound by surgeons without prior sonographic training were high, with 90% accuracy and 100% NPV in the diagnosis of partial versus nonpartial CTR (Tables 3 and 4). We found excellent internal consistency between ultrasound findings and CTR, showing excellent reliability (high precision). There was a high and significant correlation between ultrasound findings and open evaluation results, showing good to excellent validity (high accuracy) (Table 3).
Table 3.
Ultrasound in the Diagnosis of Partial Carpal Tunnel Release.
| Partial release | ||
|---|---|---|
| Positive | Negative | |
| Ultrasound test | ||
| Positive | 6 (TP) | 1 (FP) |
| Negative | 0 (FN) | 3 (TN) |
Note. TP = true positive; FP = false positive; FN = false negative; TN = true negative.
Table 4.
Diagnostic Characteristics of Ultrasound in the Diagnosis of Partial Carpal Tunnel Release.
| Reliability | |
| Cronbach α coefficient | 0.924 |
| Validity | |
| Pearson correlation coefficient | 0.869 (P = .001) |
| Sensitivity | 100% |
| Specificity | 75% |
| PPV | 86% |
| NPV | 100% |
| Accuracy | 90% |
| Sensitivity = TP/(TP + FN). Specificity = TN/(FP + TN). PPV = TP/(TP + FP). NPV = TN/(FN + TN). Accuracy = (TP + TN)/(TP + TN + FP + FN). | |
Note. PPV = positive predictive value; NPV = negative predictive value; TP = true positive; FN = false negative; FP = false positive; TN = true negative.
We observed 2 errors out of 10: one with 25% partial eCTR that was diagnosed as 50% partial release and the other with 100% nonpartial eCTR that was diagnosed with 50% partial release (Table 2).
Discussion
The aim of this study was to test the hypothesis that, with basic training, incomplete release of the carpal tunnel can be identified with a high degree of accuracy. Using a cadaveric model of incomplete CTR, we demonstrated 100% sensitivity and 86% specificity for the ultrasound used by novice sonographers for the diagnosis of incomplete CTR.
In addition to the high degree of sensitivity and specificity, these findings are notable for the high degree of agreement between ultrasound and open exploratory diagnosis of incomplete TCL release. Only one of 10 cases was deemed incompletely released by US when a complete release was performed (false positive). This indicates a higher-than-expected degree of precision for ultrasound when used by inexperienced operators.
A 2013 study by Karabay et al. recognized a 3.5-mm hypoechoic defect in the TCLs of 36 consecutive open CTR patients 2 weeks after surgery. However, all these patients were evaluated by a radiologist with extensive experience in the field.8 Similarly, in a 2015 study of ultrasound as a diagnostic tool to evaluate median nerve displacement and compression, the technique demonstrated 70% sensitivity when performed by an experienced physician.9
The findings constitute a proof-of-concept for limited use of ultrasound by orthopedic surgeons. Multiple studies of musculoskeletal ultrasound have demonstrated that user experience affects the accuracy and the reliability of the modality.11-13 Although this is traditionally used as a justification for the use of ultrasound by trained radiologists, efforts toward cost containment and efficiency have prompted calls for greater use of focused ultrasound examination by other specialists.14,15 A 2011 report by Tan et al discussed the use of ultrasound by hand surgeons in incomplete CTR to localize areas of pathology, reporting accurate ultrasound diagnosis of persistent median nerve compression in 4 patients following failed CTR.16
Clinically, screening for incomplete CTR has significant implications for treatment. Persistent symptoms after CTS can be due to multiple factors, including circumferential fibrosis around the median nerve, proliferative tenosynovitis, and amyloidosis as well as incomplete release.4,17 It has been demonstrated, however, that incomplete release as a cause of persistent symptoms is associated with better outcomes after revision surgery.18,19 An ultrasound demonstrating incomplete release thus indicates a higher likelihood of success with a revision surgery.
In a living subject, the area of failed release may be quite small. A 200-patient series of revision CTRs found incomplete release of the distal edge of the TCL in 65 patients (33%) and incomplete proximal release in 27 (14%).18 In a series of CTR revisions, Jones et al described the offending nontransected area of the TCL as being only a few millimeters in length.19 This observation brings into question the applicability of our model.
Due to the constraints of the method used to section the model carpal tunnels, it was not possible to create models with a few millimeters of remaining fibers at the distal end of the TCL. It is unclear where the detection threshold of the ultrasound evaluation lies. Anecdotally, we were able to determine complete versus partial transection of the carpal tunnel based on changes in the path of the median nerve (which takes a sigmoid course in the presence of retained ligament fibers) rather than direct visualization of the TCL. A more clinically valid evaluation of the technique, however, will require recalculating sensitivity with a more clinically valid realistic model.
Our model is also limited in that the eCTR system used transects the TCL in a distal-to-proximal manner, limiting our ability to create isolated distal areas of incomplete transection. A 2013 study of intraoperative findings in revision CTRs by Zieske et al4 found that at last 80% of incomplete releases had some proximal area of incomplete release. Only 14% of incomplete releases were incomplete at the distal end. It is unclear how many of these patients had endoscopic procedures. While the risk of proximal versus distal incomplete release depends to some extent on the endoscopic technique used, it must be acknowledged that this study does not specifically evaluate distal incomplete releases.
Another major limitation of this study is the low observer number. While the results demonstrate excellent accuracy and internal consistency, the design of the study renders it unable to assess interobserver reliability. The relatively small number of cadaveric limbs involved in the study also limits the study’s power. The study also has limitations inherent in any cadaveric model. There is no surrounding tissue reaction or healing observed that would be present in postoperative patients. This study is not designed to identify the subset of patients with recurrent (rather than persistent) CTS—that is, those patients who are relieved of symptoms following surgery and then redevelop CTS. However, given the demonstrably better outcomes of revision CTR in the setting of incomplete release,19,20 identification of a partial release has prognostic and surgical-planning implications. Ultimately, the efficacy of this technique will need to be evaluated using preoperative studies, with intraoperative findings during the revision surgery used as a criterion standard to confirm sensitivity and specificity.
This series serves as a pilot study for the in-depth evaluation of hand imaging using ultrasound in the office. Future studies will involve larger numbers of observing physicians and more physiological models.
Use of ultrasound for focused assessment of specific problems in an office setting potentially holds great value for the hand surgeon. The accuracy and reliability of diagnostic ultrasound demonstrated in this study provide additional support for its expanded use in an outpatient setting, warranting further investigation into the clinical applications of ultrasound in the hand surgeon’s office.
Footnotes
Ethical Approval: This study was approved by our institutional review board.
Statement of Human and Animal Rights: This article does not contain any studies with human or animal subjects.
Statement of Informed Consent: No consent was required for this study, as the experiments were performed on cadaveric specimens.
Declaration of Conflicting Interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
References
- 1. Sayegh E, Strauch R. Open versus endoscopic carpal tunnel release: a meta-analysis of randomized controlled trials. Clin Orthop Relat Res. 2015;473:1120-1132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Zuo D, Zhou Z, Wang H, et al. Endoscopic versus open carpal tunnel release for idiopathic carpal tunnel syndrome: a meta-analysis of randomized controlled trials. J Orthop Surg Res. 2015;10:12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kang H, Koh I, Lee T, et al. Endoscopic carpal tunnel release is preferred over mini-open despite similar outcome: a randomized trial. Clin Orthop Relat Res. 2013;471:1548-1554. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Zieske L, Ebersole GC, Davidge K, et al. Revision carpal tunnel surgery: a 10-year review of intraoperative findings and outcomes. J Hand Surg Am. 2013;38:1530-1539. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Brown MG, Rothenberg ES, Keyser B, et al. Results of 1236 endoscopic carpal tunnel release procedures using the Brown technique. Contemp Orthop. 1993;27:251-258. [PubMed] [Google Scholar]
- 6. Forman DL, Kirk H, Caulfield KA, et al. Persistent or recurrent carpal tunnel syndrome following prior endoscopic carpal tunnel release. J Hand Surg Am. 1998;23:1010-1014. [DOI] [PubMed] [Google Scholar]
- 7. Vartimidis SE, Herndon JH, Sotereanos DG. Failed endoscopic carpal tunnel release: operative findings and results of open revision surgery. J Hand Surg Br. 1999;24:465-467. [DOI] [PubMed] [Google Scholar]
- 8. Karabay N, Kayalar M, Ada S. Sonographic assessment of transverse carpal ligament after open surgical release of the carpal tunnel. Acta Orthop Traumatol Turc. 2013;47:73-78. [DOI] [PubMed] [Google Scholar]
- 9. Filius A, Scheltens M, Bosch HG, et al. Multidimensional ultrasound imaging of the wrist: changes of shape and displacement of the median nerve and tendons in carpal tunnel syndrome. J Orthop Res. 2015;33:1332-1340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Karabay N, Toros T, Cetinkol E, et al. Correlations between ultrasonography findings and surgical findings in patients with refractory symptoms after primary surgical release for carpal tunnel syndrome. Acta Orthop Traumatol Turc. 2015;49(2):126-132. [DOI] [PubMed] [Google Scholar]
- 11. Rutten MJ, Jager GJ, Kiemeney LA. Ultrasound detection of rotator cuff tears: observer agreement related to increasing experience. AJR Am J Roentgenol. 2010;195:W440-W446. [DOI] [PubMed] [Google Scholar]
- 12. Brulhart L, Ziswiler HR, Tamborrini G, et al. The importance of sonographer experience and machine quality with regards to the role of musculoskeletal ultrasound in routine care of rheumatoid arthritis patients. Clin Exp Rheumatol. 2015;33:98-101. [PubMed] [Google Scholar]
- 13. Ohrndorf S, Naumann L, Grundey J, et al. Is musculoskeletal ultrasonography an operator-dependent method or a fast and reliably teachable diagnostic tool? interreader agreements of three ultrasonographers with different training levels. Int J Rheumatol. 2010;2010:164518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Moore CL, Copel JA. Point-of-care ultrasonography. N Engl J Med. 2011;364:749-757. [DOI] [PubMed] [Google Scholar]
- 15. Nicolaou S, Talsky A, Khashoggi K, et al. Ultrasound-guided interventional radiology in critical care. Crit Care Med. 2007;35:S186-S197. [DOI] [PubMed] [Google Scholar]
- 16. Tan TC, Yeo CJ, Smith EW. High definition ultrasound as a diagnostic adjunct for incomplete carpal tunnel release. Hand Surg. 2011;16:289-294. [DOI] [PubMed] [Google Scholar]
- 17. Ebrahimzadeh MH, Mashhadinejad H, Moradi A, et al. Carpal tunnel release in diabetic and non-diabetic patients. Arch Bone Jt Surg. 2013;1:23-27. [PMC free article] [PubMed] [Google Scholar]
- 18. Stutz N, Gohritz A, van Schoonhoven J, et al. Revision surgery after carpal tunnel release—analysis of the pathology in 200 cases during a 2 year period. J Hand Surg Br. 2006;31:68-71. [DOI] [PubMed] [Google Scholar]
- 19. Jones NF, Ahn HC, Eo S. Revision surgery for persistent and recurrent carpal tunnel syndrome and for failed carpal tunnel release. Plast Reconstr Surg. 2012;129:683-692. [DOI] [PubMed] [Google Scholar]
- 20. Hulsizer DL, Staebler MP, Weiss A-PC, et al. The results of revision carpal tunnel release following previous open versus endoscopic surgery. J Hand Surg Am. 1998;23:865-869. [DOI] [PubMed] [Google Scholar]