3.1. Provocation Tests

Correct CTS diagnosis requires a thorough evaluation, including a physical examination and patient history. Provocation tests, also known as provocative maneuvers or clinical tests, are used during a physical examination to assess for the presence of symptoms or signs that may indicate CTS. These tests are designed to provoke or reproduce the symptoms of CTS, such as numbness, tingling, or pain in the hand and fingers, by manipulating the wrist and hand in specific ways. The most commonly used and cited provocation tests for CTS include:

Phalen’s Test: The patient flexes the wrists and holds them in a maximally flexed position for about 60 s. If numbness, tingling, or pain develops in the median nerve distribution (thumb, index, middle, and half of the ring finger), it may be indicative of CTS.

Tinel’s Sign: The healthcare provider taps or lightly strikes over the median nerve at the wrist. If this elicits tingling, numbness, or a shock-like sensation in the median nerve distribution, it may suggest CTS.

These methods are simple and low-cost; however, they are associated with significant limitations.

According to Kasundra, Phalen test had greater sensitivity in mild-moderate, moderate, and severe groups, while Tinel’s had greater sensitivity in mild group [26]. Conversely, Ghasemi-Rad reported that no trends were identified between testing positive for various provocative tests and the severity of CTS [14]. The reported sensitivity and specificity, as demonstrated in Table 3, is highly variable and dependent on method of examination and interpretation, and thus these tests, along with other provocative maneuvers (such as the aforementioned hand elevation test, tourniquet test, and carpal compression test), especially in isolation, have low positive predictive value [7].

3.2. NCS

NCS can help in the diagnosis of CTS by providing objective information about the function of the physiological health of the median nerve across the carpal tunnel on a quantitative basis [14].

NCS are specifically very sensitive in examining median nerve dysfunction caused by damage to the nerve and can define the degree of demyelination and axonal loss that has occurred [4]. One of the undeniable advantages of NCS is the ability to help identify dysesthesia conditions that may mimic or accompany CTS, such as cervical radiculopathy, ulnar nerve disease, polyneuropathy, brachial plexopathy, thoracic outlet syndrome, central nervous system (CNS) disorders, or other median nerve entrapment syndromes [14]. Moreover, preoperative NCS can provide an objective baseline for comparison in cases where surgical intervention is planned. This baseline can be useful in postoperative follow-up to assess treatment outcomes and identify any persistence or recurrence of symptoms [15].

Although the specificity of said method remains consistently above 80%, the sensitivity has been reported to be as low as 56% (Table 4). Even though many treating physicians and patients appreciate the additional information NCS provide in terms of confirming the diagnosis and ruling out other potential causes, as well as providing prognostic information, it has its limitations. Apart from unsatisfactory sensitivity, the disadvantages include potential discomfort for patients, and increased costs [37].

NCS base on the relative comparison of nerves. Several attempts have been made at establishing methodological criteria; however, due to insufficient literature, these findings should be confirmed by larger, multicenter collaborative efforts [38]. It has been stated that “establishing consensus on a reference standard for the diagnosis for CTS is the most important research goal in this area” [38].

If a patient does not exhibit symptoms that are consistent with CTS involving the typical distribution of the median nerve (i.e., the first three digits), performing NCS may not provide valuable information in confirming or ruling out CTS [36]. Therefore, the diagnostic accuracy depends on the clinical correlation between the patient’s symptoms and the nerve conduction findings. Similarly, normal NCS may not definitively rule out CTS. As pointed out by Sonoo et al., “it is possible to have all the symptoms of CTS with minimal or even no electrophysiological evidence of a MNW” (MNW = median neuropathy at the wrist) [39]. In case of mild CTS, results may be normal in up to one-third of the patients [40].

There are several reasons why NCS may have limitations in diagnosing CTS. Firstly, the short length of the involved nerve segment in the wrist-finger region may result in overall normal conduction velocity. Secondly, there is inter-individual variability in nerve conduction velocity, which can limit cross-nerve comparisons. Additionally, bilateral CTS may be as common as 60% of CTS cases [41], which can obscure side-to-side comparisons [4].

What’s more, the cutoff values used for defining CTS may need to be adjusted to account for differences in normal values and specific population characteristics to avoid false positives or false negatives. For example, in diabetic patients, the NCS may show slightly prolonged latencies compared to the general population even in the absence of CTS. If the typical cutoff value of 0.5 ms, is applied to diabetic patients, it may result in a higher number of false positives, as cases with relative latencies of 0.5–1.0 ms may actually be normal for this specific population. The chance of false positives is also higher for workers with hand-intensive jobs, who have vague hand symptoms. Finally, during standard NCS, small fiber axons are not examined, while pain is commonly the most distressing symptom, prompting the patients to seek medical help [39].

3.3. Imaging: USG

USG can provide real-time imaging of the soft tissues, including nerves, tendons, ligaments, and other structures in the wrist, and therefore be used as a complementary diagnostic test to assess the wrist for structural abnormalities. USG is cost-effective, widely available, noninvasive, and well tolerated by patients [14]. Therefore, it may be considered as a feasible alternative for diagnosing CTS in clinical settings as the first-line confirmatory test [13], especially where access to electrodiagnostic facilities may be limited [24].

As reported by Gervasio et al. [9] and expanded upon in Table 5, the sensitivity and specificity of USG are similar to NCS. Diagnostic accuracy of CTS based on class I studies using neuromuscular ultrasound is comparable to electrodiagnostic studies (Level A), and class II evidence consistently demonstrates the ability of USG to detect underlying structural abnormalities. Thus, “neuromuscular USG should be used to screen for structural abnormalities in the setting of suspected CTS (Level B) [12].” Lesions, which display symptoms similar to CTS, such as tenosynovitis, mass lesions, and anatomic defects, can be excluded from differential diagnosis [14].

USG can be used to quantify the degree of swelling of the median nerve in the carpal tunnel by calculating the nerve cross-sectional area (CSA). It has been suggested that this method may even be used to confirm clinical CTS diagnosis in pregnant women [42]. A normal median nerve typically has a transverse area of less than 10 mm². If the nerve’s transverse area exceeds this threshold, it may indicate swelling or compression of the nerve, which can be associated with varying degrees of nerve suffering. A transverse area between 10 and 12.9 mm² may indicate mild suffering, an area between 13 and 14.9 mm² may indicate moderate suffering, while if the transverse area exceeds 15 mm², it may indicate severe suffering of the nerve. USG has been reported to be more sensitive in detecting moderate to severe cases of CTS compared to mild cases [26]. The risk of false positives may be reduced in cases of mild suffering by measuring the difference between the maximum transverse area of the nerve in the carpal tunnel and the area of the nerve evaluated in the forearm at the level of the proximal third of the pronator quadratus muscle. For a difference above 2 mm², Gervasio et al. reported a sensitivity and specificity close to 100% [9].

According to Bang et al., the CSA of the median nerve proximally to carpal tunnel inlet and at the carpal tunnel inlet level, increases with the severity of CTS with statistically significant differences, while only the examination at the inlet was able to differentiate normal reference values from mild CTS [43]. This is in line with Park et al.’s findings, who concluded that measuring the nerve at the carpal tunnel inlet shows the highest rates of sensitivity [33]. Azami et al. also reported that the difference in CSA of the median nerve in mild, moderate, and severe CTS was statistically significant [20]; Ahmed et al. showed a strong correlation between CSA on USG and severity of CTS on NCS [44], while Kim et al. pointed out that USG provides a reliable correlation with the grade of electrodiagnostic abnormalities and clinical severity [45]. In another study, high-resolution USG, with a wrist-forearm ratio of 1.7, showed 96.1% accuracy in detecting CTS and correlated positively with median nerve conduction latency [46]. Ng et al. recommended median nerve CSA of > 14.1 mm² for mild CTS and > 19.5 mm² for severe CTS, whereas Ting et al. found optimal CSA cutoff points to discriminate between normal versus abnormal (10.5 mm²), normal/mild versus moderate/severe (13.4 mm²), and normal/mild/moderate versus severe groups (16.4 mm²).

Furthermore, Kutlar et al. reported, that Doppler USG is highly sensitive and specific (Table 5) and the results correlate strongly with CTS severity [28]. Furthermore, the utilization of Doppler USG may serve as a valuable approach in the diagnosis and evaluation of the severity of CTS, and can be considered as a dependable supplementary technique in the examination of CTS [47].

To summarize, there was a positive correlation and proportional increase between the USG measurements and the severity of CTS observed in NCS [21]. Additionally, ultrasound parameters could be used to predict both the initial treatment response and the likelihood of relapse [25]. Moreover, the presence of typical clinical symptoms, along with at least one sonographic measurement of the median nerve that displays pathological increase, can be a highly accurate predictor [27].

USG examination, however, is not without its limitations. Even though CSA measurements seem promising, there is currently no agreement on a universal cut-off value of CSA that can be used to diagnose CTS, nor is there a consensus on the location of maximum swelling of the median nerve, whether it is at the tunnel inlet, tunnel outlet, or within the tunnel itself [19]. Kasundra et al. noted that different ranges have been reported for “normal” USG parameters, with the CSA ranging from 9 mm² to 15 mm² [26]. While the measurement location at the level of the pisiform is the most common, the use of other anatomic landmarks (i.e., distal radioulnar joint, radiocarpal joint, wrist crease, intratunnel space, hook of the hamate, and distal flexor retinaculum edge) has also been reported. Furthermore, different studies have used varying reference standards for validation, with some relying on electrodiagnostic studies, while others depended on a combination of clinical examination and NCS [48].

Additionally, when dealing with advanced cases of CTS in elderly patients, it is possible for the nerve to exhibit less swelling than anticipated, given the extent of the damage. This occurs due to a decrease in the number of axons and a reduction in edema, which is replaced by fibrotic tissue. Consequently, the neural bundles tend to display greater echogenicity, and the external epineurium becomes thickened and hyperechoic [9].​ Oliveira et al., in their study, concluded that USG was unable to differentiate between the various groups based on their indicative signs and symptoms of CTS [49].

3.4. Imaging: MRI

MRI is a valuable tool both for evaluating primary nerve pathologies and assessing space-occupying lesions that may compress nerves. MRI can provide detailed information about morphological damage, inflammation, and the extent of nerve compression [14]. By visualizing the nerves directly, MRI can reveal any structural abnormalities, such as nerve thickening, enlargement, or discontinuity, which can aid in diagnosis and treatment planning. MRI excels in accurately showing the location of the lesion and determining the severity of nerve compression. This information is crucial for guiding surgical interventions. Additionally, MRI can be valuable in cases of rare pathological causes of conditions like CTS—it can accurately identify ganglia, hemangiomas, or bony deformities that can lead to nerve compression and mimic CTS symptoms, as well as localize these rare causes, enabling proper diagnosis and subsequent treatment [7, 15].

As outlined in Table 6 above, MRI seems to be a highly sensitive and specific method; however, the data available is limited. Early studies have shown that there is agreement between fractional anisotropy and the apparent diffusion coefficient values derived from MRI and the results of NCS [4]. MRI may provide information about CTS severity—it has been reported to be highly accurate at diagnosing CTS and moderately accurate at determining CTS severity. Park et al. states that measuring the nerve at the carpal tunnel inlet shows the highest rates of sensitivity and CSA of the median nerve was 15.1 mm² or larger [33]. Ng et al. suggested CSA cutoff values at > 15 mm² either proximal to or distal to the tunnel and > 19 mm² proximal to the tunnel as a diagnostic criterion for CTS in general and in severe cases, respectively, (sensitivity, specificity, and accuracy of prediction of severe CTS using for CSA > 19 mm² were 75.0%, 65.9%, and 69.6%, respectively) [31]. In their later work, median nerve CSA > 14.1 mm² to diagnose mild CTS and > 19.5 mm² for severe CTS has been recommended [32].

MRI can help identify associated pathologies that may not be evident on NCS or USG (58.97% of MRIs done as reported by Kasundra et al. [26]), which becomes especially crucial when surgical interventions, such as endoscopic procedures, are being considered. Furthermore, MRI plays a role in patients who have previously undergone carpal tunnel release surgery and subsequently develop recurrent symptoms. In such cases, MRI can be employed to assess the postoperative anatomy and identify any potential factors contributing to the recurrence of symptoms, such as residual or recurrent compressive lesions or scar tissue formation.

This method is, however, costly, time-consuming, and not readily available. Additionally, the variability in diagnostic criteria leads to variations in the profiles of patients included, impacting the accuracy of the tests [14].

Novel solutions are slowly being explored. Zhou et al. tested a deep neural network in CTS identification. The model achieved 0.63 accuracy of intersection over union and 0.17 s segmentation efficiency, even though the image training pool was relatively small (333 images) [50]. Funahashi et al. concluded that useful information about nerve morphology can be obtained from 3-D MRI [51]. As of now, these techniques remain relevant mostly as scientific research tools; however, they may become useful as complementary options once their cost-effectiveness and availability improve.

3.5. QST

QST can help in quantifying sensory abnormalities and provide additional insights into the severity and nature of nerve dysfunction in CTS. QST allows for the assessment of both large and small nerve fiber function, enabling a comprehensive evaluation of sensory abnormalities in CTS. Even though QST methods have been explored for several years, the data is still scarce (Table 7). Based on the promising preliminary results, however, this area is worth exploring to pinpoint the most effective approach among the plurality of different techniques.

An excellent meta-analysis, albeit focusing mostly on the pain detection aspect, has been recently published by Sobeeh et al. In the paragraphs related to thermal stimuli, they have reported a significant loss of warm sensation measured at the index finger and the middle finger with considerable and moderate heterogeneity, respectively. There has also been a significant loss of cold detection sensation, with moderate heterogeneity and no significant publication bias [10].

Numerous other studies showed promising results. Clarke et al. demonstrated that, after post hoc analysis, cold detection thresholds (CDT) of the severe group were significantly lower compared to the healthy subjects and the trend of CDT impairment was correlated with CTS severity. Warm perception impairment was notably more severe in the group with severe CTS when compared to the control group. Furthermore, the severe CTS group exhibited significantly greater impairment in warm perception compared to both the mild and moderate CTS groups [52]. Goadsby and Burke found clear evidence of abnormal small-fiber function, with higher thresholds for both hot and cold sensation [53]. In another study, deficits were revealed in all detection thresholds in the combined patient groups compared with healthy controls [54]. The patient groups exhibited diminished thermal and mechanical detection function within the primary pain area and dermatome when compared to healthy reference data, as reported by Tampin et al. [55]. A study by Lang et al. revealed a significant increase in thresholds for warmth, cold, and heat pain on the index finger compared to the control subjects [56].

Furthermore, as reported by Tamburin et al. overall severity of CTS sensory symptoms, as measured with the BCTQ Symptoms Severity Scale, was significantly correlated with CDT. Single CTS symptom analysis also showed a significant correlation with CDT, not with warm detection thresholds (WDT) however. This would indicate Ad- and C-fiber involvement in electrodiagnostic-negative CTS patients [57].

It is worth noting that, within the patient groups, the WDT can be higher in the affected arm compared to the unaffected arm. Similarly, within the control group, the WDT may be higher on the dominant side compared to the nondominant side. Additionally, heat hyperalgesia was more prominent in patients with unilateral CTS [58]. Age has also been shown to influence thermal detection thresholds (the magnitude is small, however) [52].

Another major component of CTS patient care is postoperative management. Kennedy et al. quantitatively tested the patients at 6 months postsurgery. They observed a moderate to large effect size in the change of QST parameters, specifically for thermal detection, thermal pain, and mechanical detection thresholds. However, there was no statistically significant change in the measures of mechanical pain. These findings indicate that the sensory phenotype derived from thermal QST is sensitive to clinically significant changes in sensory perception [59].

Even though QST aims at objectivizing the results, several researchers argue, that considerable subjective components are inextricable [7]. The patient groups involved in the studies are small and heterogenous. The data remains scarce and the available articles have variable methodology. This is especially important, as QST effectiveness may vary depending on the protocol, as exemplified by the analysis by Sobeeh et al. which revealed no difference of warm sensation at the thenar eminence or carpal area [10].

Apart from thermal stimuli, another aspect of QST is mechanical, pain, and vibration sensation. However, according to the meta-analysis of Sobeeh et al. either no significant differences between groups were shown in most of the pain modalities (i.e., cold, heat, mechanical, and pressure thresholds), or the heterogeneity and publication bias was significant [10]. This is in line with Schmid’s findings. The study revealed that patients exhibited higher thermal detection thresholds (p < 0.0001), suggesting dysfunction in C and Aδ nerve fibers, however, there were no significant differences in mechanical and thermal pain thresholds between the patient and control groups (p > 0.13). A skin biopsy was then performed and revealed a substantial reduction in intraepidermal nerve fiber density in patient group (p < 0.0001), confirming a significant impairment of small nerve fibers [60].

3.6. Summary

The questions around a true gold standard in diagnosing CTS are a fundamental challenge in evaluating the accuracy of different diagnostic measures. Clinical symptoms and signs, NCS, and surgical outcomes are three commonly used criteria, but none of them are perfect and can have false negatives and false positives.

To accurately evaluate the sensitivity and specificity of a diagnostic modality, it is important to compare it against a reference standard that is independent of that modality. For example, when evaluating the diagnostic yield of NCS for CTS, the reference standard should be based on clinical criteria, as suggested by Jablecki et al. [61]. However, the reference standard itself may have limitations and can also have false negatives or false positives, further complicating the evaluation process [39]. Additionally, there has been a lack of comprehensive head-to-head studies comparing some of approaches, which limits our ability to fully evaluate their comparative effectiveness [4].

This inherent challenge underscores the complexity of diagnosing this condition and highlights the need for a comprehensive approach that takes into consideration multiple factors, including clinical symptoms, signs, and results from various diagnostic modalities, to arrive at an accurate diagnosis.

According to Sonoo et al. in the absence of a definitive gold standard for CTS, it is advisable for clinicians to employ all reasonable diagnostic measures, including assessing symptoms and signs, as well as conducting NCS, to enhance diagnostic accuracy. By utilizing multiple diagnostic tools, clinicians can gather comprehensive information and make well-informed decisions regarding the diagnosis and management of CTS [39].

Provocation tests can serve as a first-line evaluation tool, quickly identifying potential CTS. However, they are not definitive, so they should be followed by more objective methods like NCS and USG to confirm the diagnosis.

There is a growing body of research papers claiming the effectiveness of carpal tunnel surgery without prior diagnostic investigation [62]. According to Fowler, based on their clinical experience, many surgeons operate without arranging NCS because of patient satisfaction. However, the placebo effect of the intervention is often not taken into account. Therefore, there exists a divergence of opinion between surgeons, who do not consider NCS as an essential component of their routine management for patients, and neurophysiologists, who believe that these tests provide clinical value and should be conducted in all patients with suspected CTS before surgery. Fowler then concludes that the key factor contributing to this discrepancy is the oversimplified perception of NCS for CTS as solely a binary confirmatory test, aiming to determine the presence or absence of the disease. From his experience, however, out of 21,000 patients who tested negative for CTS using standard diagnostic tests, 7% exhibited neurophysiological evidence of an alternative neurological condition that could account for their symptoms. Conversely, among patients with confirmed evidence of CTS, 5% also demonstrated indications of an additional neurological disorder.

These findings highlight the importance of identifying coexisting conditions, as they can impact the treatment outcomes for CTS or necessitate separate treatment approaches based on their individual merits [62]. Depending on the suspected condition, the use of different diagnostic techniques may be warranted. NCS can help assess the physiological damage to the median nerve, while USG can provide structural insights into nerve swelling and compression.

While MRI is excellent for visualizing nerve pathologies and associated conditions that may not be evident on NCS or USG [63], it is more costly and time-consuming. It may be recommended in situations where the patient is unresponsive to therapy and/or for research purposes. However, it does not offer sufficient additional valuable information to replace routine evaluation. It’s worth noting that some patients prefer MRI over NCS for diagnostic purposes [14]. Combining MRI with NCS or USG may be especially helpful in complex or recurrent CTS cases, where additional structural information is needed to guide treatment decisions.

In cases where the sensory symptoms of CTS are difficult to quantify using traditional methods, QST can be used to assess sensory abnormalities in more detail. This is particularly valuable for tracking changes in sensory function post-treatment or surgery and understanding the pathophysiology of CTS, especially when electrodiagnostic tests show normal results despite clear clinical symptoms.

Overall, these diagnostic methods should be viewed as complementary rather than standalone. Using a combination of clinical tests, NCS, USG, MRI, and QST, can provide a more accurate and comprehensive evaluation of CTS, helping to confirm the diagnosis, assess severity, and guide treatment options (Table 8). The suitable tools available should be primarily used to provide information required of them, rather than arbitrarily classify CTS as “confirmed” or “ruled out”.