Schwarzer et al. included data of 101 patients with language eloquent brain lesions who underwent preoperative rnTMS examination bihemispherically. Prior to rnTMS mapping, all patients performed two to three baseline runs of a picture-naming paradigm without stimulation, and only promptly and correctly named objects were retained for TMS mapping. Nine biometric factors (age, gender, baseline dataset, cognitive performance score, aphasia score, histology of lesion, affected hemisphere, location of lesion on the hemisphere, pain caused by examination) were included in the statistical analysis measuring their correlation with the incidence of errors during baseline naming as well as during rnTMS mapping.

The incidence of baseline errors correlated with aphasia (p < 0.0001) and cognitive impairment (p < 0.0001). No significant correlation was observed between most biometric factors and errors during rnTMS mapping. Factors significantly affecting the incidence of errors during rnTMS mapping were again aphasia (p < 0.023) and cognitive impairment (p < 0.038). Patients affected by those factors showed a significantly higher baseline error rate, starting at 28% error rate.

Patients with pre-existing aphasia or severe cognitive impairment did still make significantly more mistakes during rnTMS mapping than non-aphasic patients despite baseline stratification, rendering the question of whether the procedure is reliable in those patient groups. Baseline testing revealed a cut-off point at 28% error rate. Interestingly, age or pain (caused by the examination) did not bias the results ¹⁾.

Ille et al. performed multimodal language mapping in 35 patients with left-sided perisylvian lesions by using rTMS, fMRI, and DCS. The rTMS mappings were conducted with a picture-to-trigger interval (PTI, time between stimulus presentation and stimulation onset) of either 0 or 300 msec. The error rates (ERs; that is, the number of errors per number of stimulations) were calculated for each region of the cortical parcellation system (CPS). Subsequently, the rTMS mappings were analyzed through different error rate thresholds (ERT; that is, the ER at which a CPS region was defined as language positive in terms of rTMS), and the 2-out-of-3 rule (a stimulation site was defined as language positive in terms of rTMS if at least 2 out of 3 stimulations caused an error). As a second step, the authors combined the results of fMRI and rTMS in a predefined protocol of combined noninvasive mapping. To validate this noninvasive protocol, they correlated its results to DCS during awake surgery.

The analysis by different rTMS ERTs obtained the highest correlation regarding sensitivity and a low rate of false positives for the ERTs of 15%, 20%, 25%, and the 2-out-of-3 rule. However, when comparing the combined fMRI and rTMS results with DCS, the authors observed an overall specificity of 83%, a positive predictive value of 51%, a sensitivity of 98%, and a negative predictive value of 95%.

In comparison with fMRI, rTMS is a more sensitive but less specific tool for preoperative language mapping than DCS. Moreover, rTMS is most reliable when using ERTs of 15%, 20%, 25%, or the 2-out-of-3 rule and a PTI of 0 msec. Furthermore, the combination of fMRI and rTMS leads to a higher correlation to DCS than both techniques alone, and the presented protocols for combined noninvasive language mapping might play a supportive role in the language-mapping assessment prior to the gold-standard intraoperative DCS ²⁾.

nTMS and MEGI were performed on 12 subjects. nTMS yielded 21 positive language disruption sites (11 speech arrest, 5 anomia, and 5 other) while DCS yielded 10 positive sites (2 speech arrest, 5 anomia, and 3 other). MEGI isolated 32 sites of peak activation with language tasks. Positive language sites were most commonly found in the pars opercularis for all three modalities. In 9 instances the positive DCS site corresponded to a positive nTMS site, while in 1 instance it did not. In 4 instances, a positive nTMS site corresponded to a negative DCS site, while 169 instances of negative nTMS and DCS were recorded. The sensitivity of nTMS was therefore 90%, specificity was 98%, the positive predictive value was 69% and the negative predictive value was 99% as compared with intraoperative DCS. MEGI language sites for verb generation and object naming correlated with nTMS sites in 5 subjects, and with DCS sites in 2 subjects. CONCLUSION: Maps of language function generated with nTMS correlate well with those generated by DCS. Negative nTMS mapping also correlates with negative DCS mapping. In our study, MEGI lacks the same level of correlation with intraoperative mapping; nevertheless it provides useful adjunct information in some cases. nTMS may offer a lesion-based method for noninvasively interrogating language pathways and be valuable in managing patients with peri-eloquent lesions ³⁾.

Twenty patients with tumors in or close to left-sided language eloquent regions were examined by repetitive nTMS before surgery. During awake surgery, language-eloquent cortex was identified by DCS. nTMS results were compared for accuracy and reliability with regard to DCS by projecting both results into the cortical parcellation system.

Presurgical nTMS maps showed an overall sensitivity of 90.2%, specificity of 23.8%, positive predictive value of 35.6%, and negative predictive value of 83.9% compared with DCS. For the anatomic Broca's area, the corresponding values were a sensitivity of 100%, specificity of 13.0%, positive predictive value of 56.5%, and negative predictive value of 100%, respectively.

Good overall correlation between repetitive nTMS and DCS was observed, particularly with regard to negatively mapped regions. Noninvasive inhibition mapping with nTMS is evolving as a valuable tool for preoperative mapping of language areas. Yet its low specificity in posterior language areas in the current study necessitates further research to refine the methodology ⁴⁾.