Acupuncture compound anesthesia for traditional thyroidectomy: a systematic review and meta-analysis

YE Zhe (叶哲), WANG Dongcai (汪栋材), WU Haibin (吴海滨), LIN Jiwei (林基伟), SHU Jicheng (舒继承)

1 Hangzhou Linping Traditional Chinese Medicine Hospital, Zhejiang Province, Hangzhou 310000, China

2 Shenzhen Traditional Chinese Medicine Hospital, Guangdong Province, Shenzhen 518000, China

Abstract

Keywords: Acupuncture Therapy; Electroacupuncture; Acupuncture Analgesia; Thyroidectomy; Randomized Controlled Trials; Systematic Review; Meta-analysis

Thyroidectomy, a common and effective treatment when malfunction or deformation occurs in the thyroid,such as hyperthyroidism, goiter, and thyroid cancer, can be divided into lobectomy, isthmectomy, and other types based on the part removed[1].Although minimally invasive surgery and other new models have appeared in recent years, traditional thyroidectomy is still widely applied worldwide.According to statistics, over 100 000 traditional thyroidectomies are performed annually in the USA[2].

Like other surgeries, thyroid surgery leads to pain via the following mechanisms.When a stimulus activates nociceptive nerve endings, the stimulation will be transformed into electric signals and transported to the dorsal horn via fibers.Then, the nociceptive signals continue via the spinothalamic and spinoreticular tracts and ultimately reach the hypothalamus and cerebral cortex, where feelings of pain are generated[3].The effects of analgesia methods in traditional thyroidectomy are attributed to any disruption of this pathway[4].Cervical plexus block (CPB) and other anesthesia methods have been popular for many years due to their reliable effects.However, the side effects,such as constipation, can hardly be diminished, which has sparked interest in finding more anesthesia options[5].

Acupuncture compound anesthesia (ACA) is an important anesthesia modality derived from acupuncture analgesia, which has been widely used in China for over 2 000 years.Its possible mechanisms include promoting the release of neurotransmitters,inhibiting the transfer of pain, and suppressing the perception of nociceptive signals[6].Great interest in acupuncture for anesthesia purposes have developed since it was used alone for tonsillectomy[7], and its efficacy seemed inspiring in the early stages.However,this impression was not long-lasting[8-9].With the emergence of several disadvantages, including incomplete anesthesia, muscle tension, and stretch reactions, ACA was proposed[10].ACA overcomes the disadvantages of acupuncture anesthesia, improves the analgesic-sparing effects, and reduces analgesic consumption and perioperative complications in various types of surgery.However, only a few studies have focused on ACA for traditional thyroidectomy, leading us to evaluate its effectiveness in this systematic review and meta-analysis.

1 Methods

1.1 Study registration

This study was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)[11].The research has been registered in the International Prospective Register of Systematic Reviews (PROSPERO), and the registration number is CRD42021287948.

1.2 Literature search

1.2.1 Search strategies

Randomized controlled trials (RCTs) that examined ACA for thyroidectomy published between database inception and September 30, 2021 were retrieved from PubMed, Excerpta Medica Database (EMBASE),Cochrane Library, Web of Science, China National Knowledge Infrastructure (CNKI), Chongqing VIP Database (CQVIP), Wanfang Academic Journal Full-text Database (Wanfang), and China Biology Medicine Disc(CBM).The search strategy for PubMed is shown in Table 1.Searches were adjusted for each database.

Query Search term#1 Randomized controlled trial [Publication Type]#2 Controlled clinical trial [Title/Abstract] OR randomized [Title/Abstract] OR randomly[Title/Abstract] OR placebo [Title/Abstract] OR sham [Title/Abstract] OR trial [Title/Abstract] OR groups [Title/Abstract]#3 #1 OR #2#4 Acupuncture [MeSH]#5 Acupuncture [Title/Abstract] OR electroacupuncture[Title/Abstract] OR acupoint* [Title/Abstract] OR needl* [Title/Abstract]#6 #4 OR #5#7 Thyroidectomy [MeSH]#8 Thyroidectomy [Title/Abstract] OR thyroid surgery[Title/Abstract] OR thyroid operat* [Title/Abstract]#9 #7 OR #8#10 #3 AND #6 AND #9

1.2.2 Types of studies

This review included RCTs that investigated the use of ACA for traditional thyroidectomy and were published in English or Chinese.Non-RCTs, self-controlled studies,randomized crossover studies, quasi-randomized studies, case reports, expert opinions, and animal experiments were excluded.

1.2.3 Types of participants

Patients who underwent traditional thyroidectomy using ACA were included regardless of thyroid disease,age, race, and sex.Minimally invasive surgery, robotic thyroidectomy, ambulatory thyroidectomy, and other new models were not included.

1.2.4 Types of interventions

Since no standard definition of ACA exists, we defined ACA as “anesthesia with the application of acupuncture, with needle insertion”.Nonneedling techniques, such as point herbal patching, acupotomy,and transcutaneous electrical point stimulation, were excluded.

1.2.5 Types of control groups

The control group included all traditional anesthesia with placebo acupuncture, sham acupuncture, and the use of any other medications without acupuncture needle insertion or no additional treatment.

1.2.6 Outcome measures

Anesthesia effectiveness was the primary outcome.No pain or a bearable pain was defined as effective;pain needing additional drugs was defined as partly effective; pain needing anesthesia changing was defined as ineffective.We defined effective as 2 points,partly effective as 1 point, and ineffective as 0 points to get continuous variables.Secondary outcomes included the visual analog scale (VAS) score and other validated pain scale scores, vital signs, analgesic consumption,and adverse events.

1.3 Data collection and analysis

1.3.1 Selection of studies

Two investigators (YE Zhe and LIN Jiwei) screened studies independently according to the search strategy.All records were downloaded for further evaluation.The investigators independently selected eligible studies according to the inclusion criteria.Disagreements between the two investigators were settled by a third investigator (WU Haibin).

1.3.2 Data extraction and management

The following data were independently extracted from the included studies by two investigators (YE Zhe and LIN Jiwei), including general information (title,authors, year, and sample size), clinical characteristics of participants (age and gender), anesthesia method(types and details of anesthesia), interventions (type of acupuncture, acupuncture point selection, and other details of acupuncture), and outcomes.

1.3.3 Risk of bias assessment

Two investigators (YE Zhe and LIN Jiwei)independently evaluated the quality of the methodology using the Cochrane Collaboration’s bias risk assessment tool[12], including random sequence generation, allocation concealment, blinding of participants, blinding of outcome assessment,incomplete outcome reporting, selective outcome reporting, and other sources of bias.Each item was ranked as low risk, high risk, or unclear.Disagreements were resolved via consultation with a third investigator(WU Haibin).

1.3.4 Statistical analysis

We used Review Manager 5.3 to analyze the data.The outcomes were reported as the mean difference(MD) for continuous variables, with 95% confidence intervals (CI).The standardized mean difference (SMD)was used when the outcome was measured in different ways; otherwise, the weighted mean difference (WMD)was used.Statistical heterogeneity was assessed by the I² statistic and P value.We defined statistical heterogeneity as being insignificant if I2≤50%; otherwise,significant heterogeneity existed.Since ACA was performed in different forms, causing considerable clinical heterogeneity, a random effects model was applied, regardless of whether statistical heterogeneity was detected.Publication biases were assessed using funnel plots or Egger’s test.

2 Results

2.1 Study selection

Of 511 related articles retrieved in the initial search,16 studies were finally included based on the premise inclusion criteria[13-28].More details are shown in Figure 1.

2.2 Study characteristics and control interventions

Sixteen studies were included, involving 1 228 patients, 685 in the experimental groups and 543 in the control groups.All studies were published in China between 1999 and 2019, including 11 two-armed studies[15-21,23-26]and 5 multiarmed studies[13-14,22,27-28].Fifteen studies regarded acupuncture as an adjunctive therapy (noted as ACA1)[13-24,26-28].Four studies treated acupuncture as an independent anesthesia (noted as ACA2)[22,25,27-28].In the 16 enrolled studies, 1 applied manual acupuncture (MA)[26]and 16 applied electroacupuncture (EA)[13-25,27-28], where the frequency ranged from 1 Hz to 100 Hz (＞50 Hz was considered high frequency), and the duration ranged from 15 min to the entire intraoperative period.Four points were used, including Hegu (LI4), Neiguan (PC6), Futu (LI18),and Taichong (LR3).Hegu (LI4) and Neiguan (PC6) were the most commonly targeted points used in a total of 13 studies[13-14,16-18,20-25,27-28].More details are shown in Table 2 and Table 3.

For the control intervention, all enrolled studies applied CPB, and 11 papers applied premedication.Only 1 study used sham acupuncture with invalid stimulation[13].

2.3 Risk of bias

The risk of bias is shown in Figure 2 and Figure 3.All studies mentioned randomization.The random number table method was used in 3 papers[16,20-21], and the sealed envelope method was used in 1 paper[28].Regarding allocation concealment, 3 papers were at a high risk due to the open availability of the random number table they adopted[16,20-21], while 13 papers did not mention[13-15,17-19,22-28].Given the limitations imposed by the characteristics of acupuncture, blinding of personnel was not conducted in any of the included papers, and only 1 paper used sham acupuncture for blinding of participants[13].All studies had a low risk of attrition bias because there were no incomplete outcome data.For selective reporting, none of the enrolled studies conducted protocols, and 1 paper did not report all the outcomes it planned in its methods[14].Finally, there was no sufficient information to judge the risk of reporting bias.

Figure 2 Ratio of risk of bias in the included studies

Figure 3 Assessment of risk of bias in the included studies

Study Sample size Gender/case(male/female) Age/year Outcomes Adverse events MA W H 2010[13] T1: 20; T2: 20; C: 20 T1: 2/18; T2: 3/17; C: 3/17 T1: 36; T2: 42; C: 30 AE NR GAO X Q 2009[14] T1: 20; T2: 20; T3: 20; C: 20 NR 18-59 AE, AC NR ZHENG Y 2019[15] T: 45; C: 45 T: 16/29; C: 23/22 T: 42.89; C: 43.82 NRS, VS T: 3 PONV, 2 IA C: 3 PONV, 8 IA LIU Y Y 2006[16] T: 30; C: 30 T: 18/12; C:17/13 T: 39.5; C: 41.2 VAS, VS T: 1 DY, 2 HO, 4 PONV;C: 1 DY, 3 HO, 12 PONV YU M J 2019[17] T: 33; C: 33 T: 15/18; C: 12/21 T: 42.67; C: 41.93 FPS-R NR

2.4 Primary outcome of ACA1

Eight studies showed that the ACA1 group was significantly better than the control group, with low heterogeneity [SMD=0.62, 95%CI (0.40, 0.83), P＜0.0001,I2=36%] (Figure 4)[14,18-21,26-28].One study comparing the MA group to the control group showed that the MA group was significantly better than the control group[SMD=0.74, 95%CI (0.25, 1.23), P=0.003][26], and 7 studies showed that EA was significantly better than the control group [SMD=0.60, 95%CI (0.36, 0.84), P＜0.00001,I2=42%][14,18-21,27-28].More details are shown in Figure 4.

Figure 4 Forest plot of anesthesia effectiveness

Subgroup analysis for EA at different frequencies indicated that 1 study revealed no significant difference between the high-frequency group and the control group [WMD=0.00, 95%CI (-0.19, 0.19), P=1.00][14], and 6 studies showed that the low-frequency group had significantly better outcomes than the control group[WMD=0.43, 95%CI (0.30, 0.55), P＜0.00001,I2=15%][14,18-21,28].Group differences existed (P=0.0002).More details are shown in Figure 5.

Figure 5 Forest plot of high-frequency stimulation versus low-frequency stimulation

Subgroup analysis for stimulus duration showed that ACA1 for 10-30 min or during the entire operation was significantly better than the control group [WMD=0.28,95%CI (0.11, 0.44), P=0.001, I2=62%; WMD=0.55, 95%CI(0.33, 0.77), P＜0.00001, I2=0%].When the study of GAO X Q, et al[14]was removed, heterogeneity dropped to 24%, indicating the source of heterogeneity.Entire intraoperative stimulation was significantly better than 10-30 min (P=0.05).More details are shown in Figure 6.

Figure 6 Forest plot of ACA1 for 10-30 min stimulation versus entire intraoperative stimulation

2.5 Secondary outcomes of ACA1

2.5.1 Pain scales

A total of 6 studies focused on pain scales, of which 4 focused on the VAS[16,22-24], 1 focused on the numeric rating scale (NRS)[15], and 1 focused on the faces pain scale-revised (FPS-R)[17].Meta-analysis showed that the score of the ACA1 group was significantly better than that of the control group [SMD=-1.61, 95%CI (-2.61,-0.61), P＜0.00001, I2=95%].Four studies showed that the score of VAS in the treatment group was significantly better than that of the control group[SMD=-0.70, 95%CI (-1.31, -0.10), P=0.0002, I2=84%].When the study of LIU Y Y, et al[16]was removed,heterogeneity dropped to 0, indicating the source of heterogeneity.More details are shown in Figure 7.

2.5.2 Analgesic consumption

One study revealed significant differences between the 2 Hz group, the 15 Hz group, and the control group[WMD=-0.04, 95%CI (-0.07, -0.01); WMD=-0.06, 95%CI(-0.08, -0.03)], while no significant differences were found in the 100 Hz group [WMD=0.00, 95%CI (-0.03,0.03)][14].

2.5.3 Vital signs

Blood pressure was involved in 7 studies[15-16,19,21-22,24,28].Five studies mentioned systolic blood pressure (SBP)[15,19,21-22,24], and meta-analysis showed that the SBP in the ACA1 group was significantly lower than that in the control group[WMD=-18.80, 95%CI (-22.72, -14.89), P＜0.00001, I2=35%].Four studies mentioned diastolic blood pressure(DBP)[15,21-22,24], and meta-analysis showed that no significant difference existed between the ACA1 and control groups [WMD=-4.71, 95%CI (-14.93, 5.52),P＜0.00001, I2=92%].When the study of JIA Z Y[24]was removed, heterogeneity dropped to 0, indicating the source of heterogeneity.Two studies mentioned mean arterial pressure (MAP)[16,28], and meta-analysis showed that the ACA1 group was significantly lower than the control group [WMD=-13.49, 95%CI (-16.76, -10.22),P＜0.00001, I2=2%].

Heart rate was included in 7 studies[15-16,19,21-22,24,28].Meta-analysis showed that the heart rate in the ACA1 group was significantly slower than that in the control group [WMD=-13.29, 95%CI (-16.02, -10.55), P＜0.00001,I2=41%].

Three studies examined glucose[15,20,28].Meta-analysis showed that ACA1 was significantly lower than that in the control group [WMD=-1.25, 95%CI (-1.50, -1.01),P＜0.00001, I2=0%].More details are shown in Figure 8.

2.5.4 Adverse events

Adverse events were observed in 7 studies[15-16,18-19,21-22,24].There were 25 adverse events in the treatment group, involving 715 patients, and 64 events in the control group, affecting 573 patients.

2.6 Outcomes of ACA2

Meta-analysis showed that there was no significant difference between the ACA2 and control groups in anesthesia effectiveness, VAS, or vital signs (Table 4).

2.7 Publication bias

The anesthesia effectiveness was used as the outcome in 9 papers[14,18-21,25-28], and the Egger’s test suggested that no obvious publication bias existed(P=0.287).See Figure 9.

Figure 9 Egger’s test of the anesthesia effectiveness

Anesthesia effectiveness 82/82 WMD: 0.39 (-0.18, 0.97) 0.18 3 Visual analog scale 32/31 WMD: -0.26 (-7.00, 0.18) 0.25 1 Systolic blood pressure 30/30 WMD: 5.12 (-1.52, 11.76) 0.13 1 Heart rate 30/30 WMD: 4.39 (-1.62, 10.40) 0.15 1 Glucose 30/30 WMD: -0.18 (-0.39, 0.03) 0.09 1

4 Discussion

Sixteen articles involving 1 228 patients were included.All included articles examined the effectiveness of ACA for traditional thyroidectomy.Our study revealed that receiving acupuncture as an adjunctive therapy improved anesthesia effectiveness.EA was applied in 15 of the 16 included articles, and there were differences in points, frequency, stimulus duration, and other aspects.Hegu (LI4) + Neiguan (PC6)was the most popular combination of points in our study.A previous study showed that the sensory nerves of Hegu (LI4), Neiguan (PC6), and the thyroid converge and overlap in the dorsal ganglion[21]; thus, stimulation to these two points can inhibit pain from thyroid surgery.Subgroup analysis further suggested the effectiveness of low-frequency stimulation, which is believed to promote the secretion of endorphin and enkephalin[17].Subgroup analysis also revealed that stimulation from 10 min to the entire intraoperative period of EA improved the effectiveness, and the latter performed better, indicating a possible time-response relationship.Studies were divided into an entire intraoperational stimulation group and a non-entire intraoperational stimulation group ranging from 10 to 30 min.Although significant differences were observed in the 10-30 min group, the level of heterogeneity decreased quickly after excluding the study of GAO X Q,et al[14], which was the only study that adopted highfrequency stimulation.Another source of heterogeneity was the stimulus duration.There was a slight difference in the stimulus duration within the 10-30 min group;this difference indicates that no consensus has been reached regarding the appropriate stimulation duration for traditional thyroidectomy, and additional relevant studies are needed.

Regarding analgesic consumption, a higher consumption was needed to achieve the same anesthesia effectiveness in the control group since acupuncture and analgesia both contribute to anesthesia.The study of GAO X Q, et al[14]applied analgesic consumption as an index and observed that low-frequency stimulation (2 Hz and 15 Hz) reduced analgesic consumption, while no similar results were observed for high-frequency stimulation (100 Hz).However, analgesic consumption is not yet widely adopted as an outcome measurement.By reading the full text of the enrolled studies, we noticed that researchers preferred using the same dose of analgesic among groups (an extra analgesic was used only when inadequate analgesia occurred) rather than giving different doses of analgesic to different groups at the beginning, which indicated that there was no difference in analgesic consumption if inadequate analgesia did not exist.We speculate that this may be because of a lack of evidence of acupuncture anesthesia (e.g.,anesthetic potency) and hope this can be an index in future studies to collect more analgesic-saving-related information.

Pain scales, irreplaceable means of measuring anesthetic outcomes, showed significant differences between the ACA1 group and the control group.The VAS was the most selected pain scale in our study, and others included the NRS and FPS-R.Among these three scales, VAS and NRS showed reliable stability, while the former also showed the most susceptibility to pain change[29].In comparison, FPS-R is prone to bias because it requires patients’ attention and is susceptible to factors beyond pain itself (e.g., culture).Thus, we recommend VAS in the ACA for traditional thyroidectomy in future research.In our study, the VAS was applied in 4 studies, and significant heterogeneity was observed.When the only study that applied MA was removed[16], the heterogeneity dropped to 0 among the remaining three studies, which reported better scores relative to the control group.

Although whether vital signs can estimate the degree of pain is still controversial[30-31], they are still an important index to evaluate the patients’ state.When receiving thyroid surgery, patients undergo stress conditions with hemodynamic changes, which can influence blood pressure and other vital signs in a complex way.Studies pointed out that acupuncture could regulate this process in a bidirectional way to improve anesthesia management[32], which was observed in our study.However, a similar phenomenon was not observed in JIA Z Y’s study on DBP[24], which caused heterogeneity.Through reading the text, we believe this may be due to the small sample size (30 for each group).

Our study showed that the incidence of adverse events in the experimental group was much lower than that in the control group (25/715 versus 64/573).This is especially true for postoperative nausea and vomiting(PONV), inadequate analgesia, recurrent laryngeal nerve paralysis, and phrenic nerve paralysis.Previous studies revealed that postoperative acupuncture reduced the occurrence of PONV by adding endogenous βendorphin, changing the transfer of 5-HT3, and improving gastrointestinal status[33].Our study found that intraoperative acupuncture was also preventive for PONV.In addition to enhancing analgesia, we found that ACA reduced the occurrence of nerve paralysis,which may be caused by smaller analgesic consumption in the experimental group[15].

Acupuncture appeared to function as an independent means of anesthesia, though no significant differences in anesthetic effectiveness were observed compared to that of CPB.Nonetheless, we still hold a cautious attitude regarding the range of functions that acupuncture can serve; although the analgesic use of acupuncture was accepted by the WHO in 2003[34],anesthesia extends far beyond analgesia and is much more demanding.In addition, significant differences can be observed in other aspects.

Although 2 meta-analyses of ACA for thyroid surgery were performed in 2015 and 2016 in Chinese patients[35-36], it was necessary to perform the present systematic review and meta-analysis for several reasons.First, more evidence from newly published articles is available.The existing meta-analyses included articles before May 31, 2014, and 7 of 16 articles in our study were published after this date.Second, a subgroup analysis was performed based on the differences in the characteristics of the enrolled studies.Moreover, the outcomes were more comprehensive.For example,pain scales were used.Third, a protocol was registered in PROSPERO before we started this study, yielding smaller biases.Fourth, our study is the first systematic review and meta-analysis of ACA for traditional thyroidectomy in English.

There are some limitations in this study.First, all enrolled studies were from China, which may lead to biases and finite extrapolation.Second, high heterogeneity was observed, indicating differences in characteristics and interventions in the enrolled studies.Third, the performance bias derived from the blinding of participants and personnel was relatively high for only 1 of 16 studies using sham acupuncture as a placebo.Finally, the low quality and quantity of the enrolled studies may reduce the reliability of our findings.

5 Conclusion

ACA is beneficial to traditional thyroidectomy and can improve anesthesia effectiveness, moderate vital signs, and reduce the feeling of pain and adverse events when acupuncture is applied as an adjunct.However, these findings should be interpreted with caution due to the low quality and quantity of the enrolled studies.More high-quality evidence is needed to confirm our findings.

Conflict of Interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Acknowledgments

The authors received no financial support for the research, authorship, and/or publication of this article.The registration number of this study in the International Prospective Register of Systematic Reviews (PROSPERO)is CRD42021287948.

Authors’ Contributions

The study was designed by WANG Dongcai and SHU Jicheng.YE Zhe performed the study.YE Zhe, WU Haibin,and LIN Jiwei analyzed the data and drafted the manuscript.All authors read and approved the final manuscript.

Received: 9 August 2022/Accepted: 19 May 2023