Between May 2012 and May 2013, 311 patients were diagnosed with lung cancer at the Cancer Hospital of Harbin Medical University (Harbin, China). A total of 113 patients met the inclusion criteria of the present study; however, 23 patients refused participation. Thus, a total of 90 patients (age, 18–65 years), who accepted thoracotomy for the treatment of American Society of Anesthesiology grade I–III (16), TNM stage I (T1-2N0M0) (17) lung cancer, were evaluated in the current randomized, controlled trial. Eligible patients had to meet the following criteria: i) Histologically confirmed lung cancer; ii) pathologic stage I; iii) no major organ (liver, kidney or heart) dysfunction; iv) no preoperative anticancer treatment; and v) no other cancer site besides the lungs. Diagnosis of patients with lung cancer occurred using histocytological methods, biopsies were obtained using a bronchofiberscope or thoracoscope. The exclusion criteria were as follows: i) The detection of a local or systemic infection; ii) a history of chronic pain or regular opioid consumption; iii) a body mass index of >40 kg/m²; iv) evidence of body temperature disturbances; or v) they had previously experienced acupuncture/TAES therapies. The present study was approved by the Ethics Committee of Harbin Medical University, and written informed consent was obtained from all patients.

Patients were anesthetized using midazolam (0.05 mg/kg; Jiangsu Nhwa Pharmaceutical Co., Ltd., Jiangsu, China), fentanyl (0.03 mg/kg; Yichang Humanwell Pharmaceutical Co., Ltd., Hubei, China) and propofol (2.5–3.0 mg/kg; AstraZeneca, Macclesfield, UK). Injection of patients with vecuronium bromide (0.08 mg/kg; Zhejiang Xianju Pharmaceutical Co., Ltd., Zhejiang, China) facilitated tracheal intubation using a double-lumen tracheal tube (Portex; Smiths Medical, Dublin, OH, USA). The lungs were ventilated mechanically (tidal volume, 8 ml/kg; ventilator frequency, 12–14 bpm; Aestiva 5 7100; 7GE Healthcare Life Support Solutions, Madison, WI, USA). Anesthesia was maintained using propofol (4–6 mg/kg/h) and remifentanil (Yichang Humanwell Pharmaceutical Co., Ltd.). The treatment of patients with remifentanil was initiated at a rate of 0.1 µg/kg/min, followed by 0.05 µg/kg/min increments or decrements, adjusted according to hemodynamic variables. If a mean arterial pressure of <60 mmHg or a heart rate of <45 beats/min were detected for >5 min, patients were treated with 10 mg ephedrine or 0.5 mg atropine. Ringer's acetate (Hunan Kelun Pharmaceutical Co., Ltd., Hunan, China) was administered at a rate of 6–8 ml/kg/h, in order to maintain basal fluid requirements. All of the patients underwent thoracotomy (18) with a rib spreader and wound retractor (tumed-Surgical Instrument & Hospital Supplies GmbH, Tuttlingen, Germany), and without rib excision. None of the patients required a perioperative blood transfusion, and all of the patients were operated on by the same surgical team, using a consistent operative procedure.

An anesthesiologist performed TAES at the bilateral large intestine (LI) 4, pericardium (PC) 6, small intestine (SI) 3 and San Jiao (SJ) 6 acupuncture points. LI 4 is located on the dorsum surface of the hand, between the first and second metacarpal bones; PC 6 on the palmar aspect of the forearm, two ‘cuns’ above the transverse crease of the wrist, between the flexor carpi radialis and palmaris longus tendons; SI 3 on the ulnar side of the fifth digit, behind the head of the fifth metacarpal bone, the starting point of the little finger abductor muscle outer edge; and SJ 6 on the dorsal aspect of the forearm, two ‘cuns’ above wrist horizontal stripes between the ulna and radius. One ‘cun’ corresponds to the distance between the interphalangeal creases of the subject's middle finger. These acupoints were selected on the basis of the following findings from previous studies: Acupuncture at LI 4 and PC 6 was able to relieve abdominal pain, whereas it was associated with enhanced disease resistance and immunomodulation when performed at SI 3, and a reduction in incision pain associated with a rib wound when performed at SJ 6 (19–21).

The patients arrived at the laboratory at 9:00 AM on the assessment day, after which the study design and techniques were explained. The acupoints of all of the patients were swabbed with alcohol to reduce skin impedance, and were then covered with cutaneous self-adhesive electrode pads (size, 16 cm²) connected to a HANS-200 device (Han's Acupoint Nerve Stimulator; Neuroscience Research Institute, Peking University, China). The acupoints were stimulated in the standard dense-and-disperse mode for 30 min, involving alternate stimulation at 2 Hz and 100 Hz every 3 sec (frequency, 2/100 Hz). The intensity of stimulation was set at 4–12 mA to initiate minor muscle contractions. Patients were informed that they may or may not feel the current.

The subjects were randomized equally into i) thoracotomy; ii) thoracotomy and sham TAES; and iii) thoracotomy and TAES groups, using a computer-generated random list with coded sealed envelopes. Group 3 received TAES at the bilateral LI 4, PC 6, SI 3 and SJ 6 points for 30 min prior to incision, and at 20, 44, 68, 92 and 116 h following surgery. Group 2 received identical electrical stimulation to group 3; however, TAES was performed at sham points, which are located 4.0 cm obliquely superior and lateral to the LI 4, PC 6, SI 3 or SJ 6 acupoints, and are not in the meridian.

Blood samples were collected prior to surgery (basal), and at 24, 72 and 120 h postoperatively. Peripheral blood mononuclear cells (PBMCs) were isolated from venous blood using density gradient centrifugation at 400 × g for 10 min at 4°C and were analyzed by flow cytometry (FACSCalibur; BD Biosciences, San Jose, CA, USA). Plasma was stored at −70°C for subsequent measurement of the expression levels of cytokines and transcription factors.

To determine the percentages of the various T-cell subsets, PBMCs were suspended at a density of 2×10⁶ cells/ml in phosphate-buffered saline. Cells were then incubated with mouse anti-human CD4 (11-0047-42; 1:1,000 dilution), IFN-γ (12-7319-42; 1:1,000 dilution), IL-4 (12-7049-42; 1:1,000 dilution), IL-17 (12-7178-42; 1:500 dilution), and forkhead box P3 (FoxP3; 12–4777-42; 1:500 dilution) monoclonal antibodies or isotype-matched immunoglobulin G controls (12-4998-82; 1:1,000 dilution) for 30 min at 37°C. All antibodies were purchased from eBioscience, Inc. (San Diego, CA, USA). For intracellular IL-17A staining, the cells were treated with PMA (Sigma-Aldrich, St. Louis, MO, USA) at 50 ng/ml and ionomycin (Sigma-Aldrich) at 1 µM in the presence of GolgiStop (BD Pharmingen, San Diego, CA, USA) for 4 h. For the FoxP3 analysis, the cells were not stimulated. The stained cells were analyzed using a FACScan Cytometer equipped with CellQuest 6.0 software (BD FACSAria; BD Biosciences).

The protein expression levels of IL-2, IFN-γ, IL-4, IL-10, IL-17 and transforming growth factor (TGF)-β, were measured using an ELISA (eBioscience, Inc.). All of the samples were analyzed in duplicate. The minimum detectable cytokine concentration was 10 pg/ml for IL-2, 20 pg/ml for IL-4, 2 pg/ml for IFN-γ, IL-10 and IL-17, and 60 pg/ml for TGF-β.

Total RNA was extracted from the PBMCs using the TRIzol® extraction kit (Invitrogen Life Technologies, Carlsbad, CA, USA). Total RNA (2 µg) from each sample was reverse transcribed to cDNA in a 40 µl reaction mixture which contained 8 µl 5X reverse transcriptase moloney murine leukemia virus (M-MLV) buffer, 1 µl DTT, 2 µl dNTP mixture, 2 µl random primers, 1 µl reverse transcriptase M-MLV, 1 µl RNAsafe all Takara Biotechnology Co., Ltd., Dalian, China) and 23 µl RNase free ddH₂O. TaqMan primers and probes targeting human T-bet, GATA binding protein 3 (GATA3), RAR-related orphan receptor (ROR)-γt, and FoxP3 were used. Samples (2 µl cDNA) were analyzed using PrimeScript™ RT-PCR kit (Takara Biotechnology Co., Ltd.) and the ABI Prism 7900 Sequence Detection System (Applied Biosystems Life Technologies, Foster City, CA, USA). The PCR cycling conditions were as follows: 94°C for 3 min, followed by 94°C for 15 sec, 55°C for 30 sec and 72°C for 40 sec for 40 cycles, and final extension at 72°C for 10 min. The primer pairs used were as follows: Forward, 5′-GCTGGAGAAAAGAAGACAAGAAAG-3′ and reverse, 5′-AAGAAAAAACACACACCCACACAC-3′ for T-bet (496 bp); forward, 5′-AGGGAGTGTGTGAACTGTGGG-3′ and reverse, 5′-CTTCGCTTGGGCTTAATGAGG-3′ for GATA3 (253 bp); forward, 5′-GCAATGGAAGTGGTGCTGGTT-3′ and reverse, 5′-AGGATGCTTTGGCGATGAGTC-3′ for RORγt (192 bp); and forward, 5′-CACGCATGTTTGCCTCTTCAGA-3′ and reverse, 5′-GTAGGGTTGGAACACCTGCTGGG-3′ for FoxP3 (235 bp). The mRNA expression levels of the target genes were normalized against GAPDH, and relative mRNA expression levels for each cytokine were calculated. Relative fold changes of gene expression were calculated using the ∆∆Cq method using ABI Prism® 7900HT software, and the values were expressed as 2^−∆∆Cq (22), relative fold changes of target gene expression were normalized against GAPDH.

Postoperative pain at rest and whilst coughing were assessed using a visual analogue scale (VAS), between 0 (pain free) and 10 (worst possible pain), prior to surgery (basal), and at 2, 6, 12, 18, 24, 36, 48, 60, 72, 96 and 120 h postoperatively. Intravenous treatment with 100 mg tramadol (Grünenthal GmbH, Aachen, Germany) was used for postoperative pain rescue, as required. Analgesic requirements, adverse events and durations of hospitalization, were also recorded.

Of the 90 patients, nine were excluded: Three due to serious postoperative complications (two in group 2, and one in group 3), and six due to incomplete data collection (three in group 1, one in group 2 and two in group 3). Thus, a total of 81 patients were included in the data analysis. There were no significant differences in the demographics and surgical information among the three groups (P>0.05; Table I).

Compared with group 1, VAS scores at rest and during coughing decreased in group 3 between 6 and 60 h following surgery (P<0.05. VAS scores between groups 1 and 2 were not significantly different (P>0.05; Fig. 1). Compared with group 2, VAS scores at rest and during coughing decreased in group 3 between 6 and 60 h following surgery (P<0.05). These results indicted that TAES at sham acupoints did not have any effect on the human body.

Representative binding patterns of the various CD4⁺ T cell subsets are presented in Fig. 2. The percentage of Th₁ (intracellular antibodies against IFN-γ, 14.4%) and Th₁₇ (intracellular antibodies against IL-17, 2.4%) cells in group 3 were significantly higher 72 h post-operation, as compared with groups 1 and 2 (9.2%; P<0.05; Fig. 2A and B). The percentage of Th₂ cells (intracellular antibodies against IL-4, 3.8%) in group 3 were significantly lower, as compared with in group 1 (8.3%) 72 h post-operation (P<0.05; Fig. 2A). There were no differences in the percentage of T_reg cells (intracellular antibodies against FoxP3) among all three groups (P>0.05; Fig. 2C).

Protein expression levels of IL-2 and IFN-γ were significantly increased at 72 and 120 h postoperatively in group 3, compared with group 1 (P<0.05; Fig. 3A and B). By contrast, IL-10 protein expression levels were significantly decreased at 72 h postoperatively in group 3, compared with group 1 (P<0.05; Fig. 3C). IL-17 protein expression levels were increased 120 h postoperatively in group 3, compared with group 1 (P<0.05; Fig. 3D). There were no significant differences in the protein expression levels of IL-4 and TGF-β among the three groups (P>0.05; Fig. 3E and F, respectively).

T-bet mRNA expression levels were significantly upregulated at 24, 72 and 120 h postoperatively in group 3, compared with group 1 (P<0.05; Fig. 4A). RORγt mRNA expression levels were significantly upregulated at 72 h postoperatively in group 3, compared with group 1 (P<0.05; Fig. 4B). Conversely, GATA3 mRNA expression levels were downregulated at 72 and 120 h postoperatively in group 3, compared with group 1 (P<0.05; Fig. 4C). There were no significant differences in FoxP3 mRNA expression levels among the three groups (P>0.05; Fig. 4D).

Rescue analgesic demands were significantly lower in group 3, compared with group 1 (P<0.05). In addition, the incidence of postoperative nausea/vomiting and infection was significantly less in group 3, compared with group I (P<0.05; Table II). Hospital stays in group 3 (8.7±1.9 days) were marginally shorter, compared with groups 1 and 2 (10.3±1.5 and 11.5±1.6 days, respectively); however, there were no significant differences among the three groups (P>0.05).