Journal Logo

Medicine Correspondence Blog

The Medicine Correspondence Blog allows authors to post Letters to the Editors, Reviews, and other editorial writings that are not considered original research.

Friday, January 3, 2020

Author Response to Letter to the Editor: Acupuncture and weight loss in Asians, a PRISMA-compliant systematic review and meta-analysis

We are very grateful to the rigorous and careful professor for raising the questions.

We corrected the similar questions and published in Medicine (Baltimore), 2019 Dec;98(49):e18400. doi: 10.1097/MD.0000000000018400. “Acupuncture  and weight loss in Asians: A PRISMA-compliant systematic review and meta-analysis: Erratum.”

The authors questions are in three aspects.

1. The first question is about the correlation coefficient.

Response: 

If the change of mean and standard deviation from baseline to the endpoint of the trial are not reported and we have to estimate. The correlation coefficient, not a fixed number, is important for the estimated standard deviation, which can have affect on the results and is not easy to determine. We made the decision by referring to several articles [1,2] and assumed the correlation coefficient to be 0.4.
1. Zhang F, Tong Y, Su N, Li Y, Tang L, Huang L, Tong N. Weight loss effect of glucagon-like peptide-1 mimetics on obese/overweight adults without diabetes: A systematic review and meta-analysis of randomized controlled trials. J Diabetes,2015,7(3):329-339.
2. Zhang F, Tang L, Zhang Y, Lü Q, Tong N. Glucagon-like peptide-1 mimetics, optimal for Asian type 2 diabetes patients with and without overweight/obesity: meta-analysis of randomized controlled trials. Sci Rep.2017,7(1):15997. 

2. The second question is about the discrepancies in mean differences reported with original manuscripts

“For the original paper of Hsu 2009 [3], the means of differences for the control group are -0.3 for BMI and -2.3 for WC, where in Junpeng’s work [1], these were reported as 0.3 and 2.3 respectively.”

Response: 

This was our mistake. We are very sorry for it. We revised the data and recalculated. 
The overall efficacy of acupuncture relative to control treatment was evident from a significant difference in the reduction of BMI (WMD -1.20kg/m2; 95% CI -1.91, -0.48) (Figure 1). In the subgroup analyses, significant differences in the reduction of BMI were noted between acupuncture and sham acupuncture (WMD -0.79kg/m2; 95% CI -0.99, -0.59).

The overall efficacy of acupuncture relative to control treatment was evident from the significant difference in the reduction of waist circumference (WMD -1.85cm; 95% CI -3.20, -0.49) (Figure 2). In the subgroup analyses, there were significant differences in the reduction of waist circumference between acupuncture plus diet and exercise, and diet and exercise (WMD -4.35cm; 95% CI -6.16, -2.54), and acupuncture and no intervention (WMD -0.29 cm; 95% CI -0.54, -0.05). There was no significant difference between acupuncture and sham acupuncture (WMD -1.28cm; 95% CI -3.96, 1.41), acupuncture plus exercise and exercise (WMD -1.07 cm; 95% CI -4.29, 2.16). 


1.png

Figure 1. Body mass index (BMI): acupuncture vs control.



2.png

Figure 2. Waist circumference: acupuncture vs control.


3. The third question is about the missing data.

“In addition, the outcome of waist circumference from Yeo 2014 [4] appear to be missing from the meta-analysis, even though these statistics are reported in the original paper. ”

Response: 

We omitted the data. This was our mistake and We feel shamed for it. We added the data and recalculated. 
The overall efficacy of acupuncture relative to control treatment was evident from the significant difference in the reduction of waist circumference (WMD -1.85cm; 95% CI -3.20, -0.49) (Figure 2). In the subgroup analyses, there were significant differences in the reduction of waist circumference between acupuncture plus diet and exercise, and diet and exercise (WMD -4.35cm; 95% CI -6.16, -2.54), and acupuncture and no intervention (WMD -0.29 cm; 95% CI -0.54, -0.05). There was no significant difference between acupuncture and sham acupuncture (WMD -1.28cm; 95% CI -3.96, 1.41), acupuncture plus exercise and exercise (WMD -1.07 cm; 95% CI -4.29, 2.16). 


3.png

Figure 2. Waist circumference: acupuncture vs control.


Combined with the questions raised, we carefully verified the data in the article, and made CORRECTIONS in red as follows.

CORRECTIONS:

1. Page 1

Abstract

Results: Twelve RCTs involving 1151 subjects were included. Compared with the control groups, the acupuncture groups exhibited significantly greater reductions of body mass index (BMI) (WMD -1.20kg/m2; 95% CI -1.91, -0.48)and waist circumference (WMD -1.85cm; 95% CI -3.20, -0.49) In the subgroup analyses, significant differences in the reduction of BMI was observed between the acupuncture and sham acupuncture groups, the acupuncture plus diet and exercise, and the diet and exercise groups, and the acupuncture and no intervention groups, but not between the acupuncture plus exercise and exercise groups.

2. Page 3

3.3. Effect of acupuncture on BMI

The overall efficacy of acupuncture relative to control treatment was evident from a significant difference in the reduction of BMI (WMD -1.20kg/m2; 95% CI -1.91, -0.48) (Fig. 4). In the subgroup analyses, significant differences in the reduction of BMI were noted between acupuncture and sham acupuncture (WMD -0.79kg/m2; 95% CI -0.99, -0.59), acupuncture plus diet and exercise and diet and exercise (WMD:-2.27kg/m2; 95%CI: -4.26, -0.29) and acupuncture and no intervention(WMD:-1.70kg/m2;95%CI: -2.59, -0.81). No significant differences were observed in the comparisons of acupuncture with placebo acupuncture (WMD:-0.98kg/m2;95%CI: -2.26, 0.30), acupuncture plus laser acupuncture with laser acupuncture (WMD:-0.04kg/m2;95%CI: -1.21, 1.13) and acupuncture plus exercise with exercise (WMD:-0.50kg/m2;95%CI: -2.20, 1.20).


4.png

Figure 4. Body mass index (BMI): acupuncture vs control.


3. Page 3 and Page 5:

The overall efficacy of acupuncture relative to control treatment was evident from the significant difference in the reduction of waist circumference (WMD -1.85cm; 95% CI -3.20, -0.49) (Fig. 5). In the subgroup analyses, there were significant differences in the reduction of waist circumference between acupuncture plus diet and exercise, and diet and exercise (WMD -4.35cm; 95% CI -6.16, -2.54), and acupuncture and no intervention (WMD -0.29 cm; 95% CI -0.54, -0.05). There was no significant difference between acupuncture and sham acupuncture (WMD -1.28cm; 95% CI -3.96, 1.41), acupuncture plus exercise and exercise (WMD -1.07 cm; 95% CI -4.29, 2.16). 


5.png

Figure 5. Waist circumference: acupuncture vs control.


On behalf of all authors, we apologize for these errors and any inconvenience caused to the readers by these changes.


Author Correspondence

Lin Zhang, The Third Affiliated Hospital of Chengdu University of Traditional Chinese Medicine, Diabetes Mellitus Prevention and Control Center of Sichuan Province, China

Email: zhanglhx@cdutcm.edu.cn


Ying Li, Chengdu University of Traditional Chinese Medicine, China 

Email: liying@cdutcm.edu.cn