The primary objective is to assess the feasibility of the study design. Feasibility outcomes will include (1) participant recruitment and (2) participant retention for the postsurgical assessment. On the basis of past records of the thoracic surgery division at SJH, we aim to recruit 3 to 4 patients per month over a 10-month study period. We will consider monthly recruitment of 3 to 4 patients and >90% retention to be feasible, monthly recruitment of 1 to 2 patients and 85% to 90% retention as feasible with appropriate modifications, and no patients recruited in 5 out of 10 months and <85% retention as not feasible. These cutoffs will be used to evaluate the need for protocol revisions before conducting a large-scale study.

In the presurgical assessment, information regarding each participant’s age, sex, history of prior surgery, history of preexisting pain, reason for upcoming surgery, other comorbidities, adjuvant therapies, and pain medication use will be documented. In the postsurgical assessment, details of each patient’s surgery type, number of ports, adjuvant therapies, and use of pain medications will be documented by the experimenter. In addition, the anesthesia protocols used for each participant’s surgery will be collected through electronic chart data, including the anesthetic technique (volatile or total intravenous), regional anesthesia type (thoracic epidural, paravertebral, erector spinae plane, or intercostal block), delivery mode (single-shot or continuous catheter), and duration of postoperative infusion.

This information will be reported descriptively and considered during data analysis so that its potential impact can be considered in the interpretation of the results.

The questionnaires that will be completed in a self-administered manner during each assessment session are presented in Textbox 1.

TMS will be performed using a 50-mm figure-of-8 coil attached to a Magstim 200² stimulator (Magstim). This study will focus on stimulating the motor hot spot of the right first dorsal interosseus (FDI) muscle for all TMS measures. Surface electrodes (9 mm Ag-AgCl) will be used to record electromyography (EMG) activity from the FDI muscle of the right hand.

Resting motor threshold, defined as the lowest intensity required to evoke a motor evoked potential (MEP) ≥50 μV in 50% of the time, will be determined for each participant’s FDI muscle [37]. Active motor threshold, defined as the lowest intensity required to evoke an MEP≥200-μV in 5 out of 10 consecutive trials [37] while participants maintain a sustained isometric contraction of the right FDI muscle to 10% of their maximum voluntary contraction (MVC), will also be recorded. This MVC value will be determined beforehand, via participants performing a brief 5-second isometric voluntary maximum contraction of the right FDI muscle, with maximal EMG activity recorded.

MEPs will be acquired by stimulating the motor hot spot for the right FDI muscle at 120% of the resting motor threshold for a total of 30 pulses at rest. MEPs will also be collected during voluntary contraction of the target FDI muscle to 50% MVC, with 20 pulses delivered at an intensity that evokes a resting MEP with approximately 1 mV peak-to-peak amplitude (determined beforehand). Average MEP amplitude will be calculated for each condition, and this measure will be used to assess the levels of corticospinal excitability.

Short-latency intracortical inhibition (SICI), a measure of cortical inhibition, is measured using paired-pulse TMS with an interstimulus interval of 2 ms between the conditioning stimulus (CS) and test stimulus (TS). The CS will be set to 90% of the active motor threshold, and the TS will be set to the intensity evoking an MEP with a peak-peak amplitude of approximately 1 mV in the right FDI. SICI will be calculated as the ratio of the average conditioned MEP (CS-TS) to the amplitude of the average unconditioned MEP (MEP-TS only), with smaller values indicating greater levels of intracortical inhibition, and vice versa. A total of 25 TS and 25 CS+TS pulses will be delivered.

Cortical silent period (CSP), a measure of spinal and cortical inhibition, is an interruption in the ongoing voluntary EMG signal that occurs following an MEP evoked by a suprathreshold TMS pulse. The CSP will be probed by delivering single-pulse TMS during voluntary contraction at approximately 50% MVC of the FDI. A total of 20 pulses will be delivered at an intensity evoking a resting MEP with a peak-peak amplitude of approximately 1 mV (determined beforehand). CSP duration will be measured for each trial and averaged. CSP onset will be defined as the beginning of the TMS-evoked MEP, and CSP offset will be defined as when the voluntary EMG signal reappears. Once CSP onset and offset are determined, CSP duration will be calculated.

Quantitative sensory testing (QST) will be used in this study to assess somatosensory function before and after surgery. The series of tests to be used includes the mechanical detection threshold (MDT) and the pressure pain threshold (PPT) tests. For patients undergoing lobectomy, wedge resection, or segmental resection, QST will be performed unilaterally at 3 sites on the sides of the trunk (approximately the fourth, seventh, and ninth intercostal spaces) [38,39]. For patients undergoing MIE, QST will be performed bilaterally at 3 sites on the sides of the trunk (approximately the fourth, seventh, and ninth intercostal spaces) and 1 site at the abdomen (around the navel) [40]. QST results will be analyzed in their original measurement units (lb for PPT and mN for MDT).

EEG will be recorded using an active EEG electrode system (Quik-Cap Neo Net; Compumedics Neuroscan) connected to a Grael 4K EEG amplifier (Compumedics Neuroscan) with 32 channels of recording electrodes positioned across the cortical hemisphere using the international 10 to 20 system. EEG data will be used to derive the following 3 outcomes.

Spontaneous EEG will be recorded under 2 conditions: eyes open (EO) and eyes closed (EC). Each condition will last 5 minutes. During the EO condition, a white fixation cross will be presented at the center of the screen for patients to focus on. During the EC condition, patients will sit relaxed with their eyes closed. Experimenters will ensure that patients remain awake during the EC condition through continuous monitoring.

Following findings from Vuckovic et al [41], features will be based on the power spectral density calculated for each frequency band (alpha [8-12 Hz] and beta [13-30 Hz]) for the EO and EC tasks. Additionally, the ratio between the EC and EO states for each frequency band will be calculated [41]. Features will be computed for each of the 32 channels.

Event-related desynchronization (ERD) will be measured during periodic submaximal isometric contractions of the abductor pollicis brevis muscle.

ERD measures the reduction in the power of synchronized neural oscillations across various frequency bands in response to movement and serves as a marker of increased cortical excitability [28]. ERD will be assessed using an EEG recorded from electrodes positioned over the primary motor cortex (C3 and C4). ERD will be calculated by comparing alpha and beta band power during the active movement phase (0.5-1.5 seconds) to a premovement baseline period (−3.5 to −0.5 s), using standard event-related spectral perturbation analysis [42].

Corticomuscular coherence (CMC) will be measured during periodic submaximal isometric contractions of the abductor pollicis brevis muscle.

CMC is a measure of the synchrony between oscillations in EEG signals and EMG signals at the muscle, suggested to reflect reciprocal communication between the motor cortex and active muscle during movement [43]. Deficits in CMC have been shown to reflect impairments in sensorimotor integration in chronic pain [44]. In this study, CMC will be calculated from EEG recordings over C3 and C4 and EMG signals from the right abductor pollicis brevis muscle using established correlation-based methods [45], focusing on the mu (8-12 Hz) and beta (15-30 Hz) frequency bands.

The measurements acquired during each assessment are summarized in Table 1. Each of these TMS and EEG measures used has shown good to excellent within-session or between-session reliability [46-51], and therefore provides a reliable basis for evaluating cortical activity, excitability, and inhibition.

At T1, participants will be categorized as CPSP+ or CPSP–. CPSP– will be defined as a participant experiencing no persistent pain related to the surgery 3 months after the operation. CPSP+ will be defined as the presence of new persistent pain that is localized to the operated area (of any severity) or pain of increased severity (if there was pre-existing pain in that site) 3 months after the operation, which cannot be attributable to any other identifiable cause. This diagnosis will be evaluated by asking the following question: Do you have persistent pain (that started or increased with or after surgery) at or near the surgical area? For patients who answer yes, the severity of pain localized to the surgical area will be assessed using the numeric rating scale [52].

The sample size for this study is based on evaluating feasibility for a larger study, and thus is not sufficiently powered to detect clinical differences. We based our sample size on pilot trial sample size calculations using a CI approach suggested by Thabane et al [53]. In addition, our sample size was determined in consultation with the thoracic surgery team at SJH, who provided data on the number of surgeries performed within the department.

Univariable logistic regression will be conducted to explore the associations between preoperative and perioperative characteristics and CPSP status. These analyses will be considered hypothesis-generating to inform the planned larger cohort study. We will test the individual association of the following independent variables with the diagnosis of CPSP: demographic and clinical information (age, sex, baseline pain intensity, depression and anxiety levels, surgery type, and regional anesthesia information), TMS measures (SICI, CSP, and MEPs), EEG measures (band activity, ERD, and CMC), and QST measures (MDT and PPT). Variables that show a statistically significant relationship (P<.05) will be considered independent predictors of CPSP development in this pilot study. Odds ratios with 95% CIs will also be reported.

Exploratory comparisons of postsurgical TMS and EEG measures between patients who are CPSP+ and CPSP– will be conducted using 2-sample 2-tailed t tests. Significance will be set to α=.05, and effect sizes (Cohen d) will be reported with 95% CIs. Before this analysis, normality will be assessed with the Shapiro-Wilk test, and outliers will be removed with Grubbs test.

These secondary exploratory aims will be analyzed using an available-case approach, whereby participants with missing data for a given variable will be excluded only from analyses involving that variable and retained in all others.