Authors' reply re: Safety of acupuncture during pregnancy: a retrospective cohort study in Korea

Abstract

cohort study of its kind, and the authors concluded that acupuncture may be a safe therapeutic option for relieving discomfort in pregnancy. Although a study like this is much needed and provides essential information, we noticed several problems that should be addressed before any firm conclusion can be drawn. First, the authors reported that the incidences of preterm delivery were 8.45% and 6.93% in the acupuncture and control groups, respectively. Although comparison by chi-square test did not indicate significant differences, it should be noted that the P-value and lower confidence limit approached significance in both crude and adjusted analyses. Considering the retrospective nature of this study, a small number of errors during data collection could dramatically change the results, especially at the statistical margins. It would be advisable for the authors to add the absolute difference between groups with a 95% confidence interval. Second, the authors stated that women who received acupuncture therapy were frequently diagnosed with such ailments as functional dyspepsia, lower back pain and influenza, but the physical status of women in the control group remained unclear. As a result of the potential adverse effects of these disorders on pregnancy outcomes, a direct comparison between these two populations could be affected by the patient characteristics and so fail to fully reflect the effects of the acupuncture treatment itself. Notably, the women in the acupuncture group had significantly more visits to obstetrics and gynaecology specialists, suggesting that the comparable delivery outcomes might be to the result of more frequent monitoring and more prompt treatment than in the control group. We question why the number of visits was not adjusted as a confounder when other baseline features were all included in multivariable regression analysis. Important variables are also missing from the report. We refer to gravidity, parity, and previous history of preterm birth and stillbirth, all of which are very closely associated with the outcomes that the research team chose to analyse and so should have been taken into account for adjustment. Finally, we are interested in what effect acupuncture in pregnant women would have on babies apart from similar risks in abnormal delivery, such as neonatal birthweight and major congenital malformations. Given the benefits of acupuncture on maternal symptoms, a comprehensive safety validation for neonates would further promote the widespread use of this complementary therapy. In summary, although this study is meaningful, the conclusion would be more solid if more accurate, detailed and complete analyses were performed.&