The preventive effect of dexmedetomidine on adverse reactions induced by uterotonics in cesarean section: a meta-analysis

Background: To evaluate the preventive effect of dexmedetomidine(DEX) on adverse reactions induced by uterotonics used in cesarean section. Methods: Randomized controlled trials(RCTs) comparing dexmedetomidine with placebo on preventing the side effects of uterotonics in cesarean section were searched in PubMed, The Cochrane Library, and CNKI. Results: Fifteen articles with 1174 patients were included in this study. Compared with placebo, DEX signicantly reduced the incidence of adverse drug reactions, such as nausea, vomiting, facial ushing, chest distress, hypertension and tachycardia ([OR = 0.12, 95% CI (0.09, 0.17), P < 0.00001], [OR = 0.12, 95% CI (0.09, 0.18), P < 0.00001], [OR = 0.21, 95% CI (0.16, 0.29), P < 0.00001], [OR = 0.19,95%CI (0.12,0.30), P < 0.00001], [OR=0.09, 95%CI (0.05,0.14), P < 0.00001], [OR=0.04, 95%CI (0.02,0.07), P < 0.00001]),which were induced by uterotonics used in cesarean section. Conclusion: Dexmedetomidine can prevent adverse reactions including nausea, vomiting, facial ushing, chest distress, hypertension and tachycardia caused by uterotonics used in cesarean section.


Background
Postpartum hemorrhage is a serious complication of pregnant women and also the leading cause of maternal death worldwide. The four major causes of postpartum hemorrhage are uterine fatigue, birth canal damage, placental factors and coagulation dysfunction, of which uterine fatigue is the most common one, accounting for 70%~90% [1] . And hemorrhage caused by uterine fatigue is much more common in caesarean section than natural delivery. WHO guidelines on prevention and treatment of postpartum hemorrhage also clearly points out that all of the parturients should use uterotonics prophylaciously in the third stage of labor, which can prevent and reduce the occurrence of postpartum hemorrhage [2] .At present, the commonly used uterotonics in clinical practice include oxytocin, carboprost tromethamine and carbetocin, etc. However, these agents have a series of adverse reactions, including gastrointestinal discomfort(nausea and vomiting) and cardiovascular side effects(increased blood pressure, increased heart rate, chest tightness, facial ushing) [3] .
These adverse reactions usually have a bad impact on surgical operation, anesthesia safety and postpartum recovery. Dexmedetomidine(DEX), a new type of highly selective alpha-2 adrenergic agonist with sedative, analgesic and anti-anxiety effects, has been widely used in cesarean section these days. Studied has been reported that it can reduce the incidence of nausea, vomiting and chilla during spinal anesthesia in cesarean section [4,5,6,7] . And in recent years, there is a increasing number of randomized controlled trials (RCTs) involving the use of DEX in cesarean section with the use of uterotonic agents. So the aim of this study is to explore and evaluate the effect of dexmedetomidine in the prevention of side effects of uterotonic in cesarean section through meta-analysis and to provide evidence-based basis for the rational use of drugs for obstetric surgery.
This meta-analysis was written according to the preferred reporting items for systematic reviews and PRISMA statement and a drafted study protocol was also made which was consistent with the Cochrane Collaboration format.

Inclusion criteria
(1)Study design:only randomized controlled trials (RCTs) were considered in this study.
(2)Participants: pregnancy patients undergoing scheduled cesarean section, with American Society of Anesthesiologists (ASA) physical status I-II.
(3)Intervention: placebo(saline) was used in the control group and dexmedetomidine was used in the observation group.

Exclusion criteria
(1) non-RCTs; (2) Reviews; (3) Animal experimental research and in vitro experimental research (4) Unclear methods of drug administration; (5) No intervention of uterotonics in cesarean section; (6) Dexmedetomidine used in combination with other drugs in treatment; (7)No data provided for the calculating of adverse reactions.

Search strategy
Pubmed, Cochrane Library,Embase,CNKI and other Chinese databases were searched to indentify the clinical trials related to dexmedetomidine used in cesarean section. The retrieval period was from the establishment of each database to June 2019 and the search terms are "Dexmedetomidine" and "Cesarean". Titles, abstracts, and full texts of relevant articles were screened after the excluding of duplicated ones.

Study selection and data extraction
Two independent researchers screened the literatures strictly according to the inclusion and exclusion criteria and extracted relevant information and data from the included literatures independently. The information mainly included basic information, research objective, interventions and outcome indicators. The two researchers cross-checked and discussed it after completion.
Accessing of Literature quality Literature quality was accessed by two researchers separately according to Jadad scale. The evaluated items included the following four: (1) whether it was randomized and whether the method was described and correct; (2) whether blinding was performed and whether the method was correct(3) whether allocation concealment was implied and whether the method was correct;(4) whether the data were complete. Literature with a score bellow 4 was considered of low quality.

Statistical methods
Data analysis was performed using RevMan5.3 statistical software provided by Cochrane Library(The Cochrane Collaboration, The Nordic Cochrane Center, Copenhagen, Denmark). In our study, incidence of side effects belongs to categorical outcomes, so the odds risks (ORs) and 95% Con dence Interval(CI) were calculated. Heterogeneity test or Q-test was adopted for heterogeneity analysis. If I 2 < 50%, the data was not signi cantly homogeneous, xed effect model was used; otherwise, random effect model was selected. The results of meta-analysis described in forest maps.

Results of literature searching
A total of 1093 articles were initially identi ed and 611 literatures were retained after excluding duplicate ones. 127 articles remained through screening of titles and abstracts. Reviews, non-clinical trial studies and those that did not meet the objective of our study were discarded. The 127 studies were screened in full text according to the inclusion and exclusion criteria. 100 articles was excluded in which the participants were not in conformity with the requirements, the uterotonic agents was not applied, saline was not used for control group or the use of DEX was in combination with other drugs. After the ruling out of studies in which the outcome index was unmatched or the data was incomplete, nally a total of fteen articles remained. The ow chart of study selection process is shown in gure 1.

Figure1. Flow diagram of the study selection process
The basic information of the fteen studies is shown in table 1. A total of 1174 patients were included, including 589 cases in the observation group (DEX) and 585 cases in the control group (saline).All the subjects were given uterotonic agents (oxytocin, caprostatin trimethamine, or carbetocin) after delivery.

Quality evaluation of the articles
The included fteen articles were all RCTs, in which four trials didn't explain the concrete randomized method [8,9,21,22] , one paper adopted the method of draw [15], one piece applied computer random sequence [17] , the remained took random number table method [10][11][12][13][14]16,[18][19][20] . Blind method was mentioned in two of the research [14,17] . And the detail information of allocation concealment, cases of exit and lost to follow-up were reported in none of the fteen trails. The quality scores of the included literatures were shown in table 2. As we can draw from it, two of the trials were graded 4 scores, nine 3 scores and four 2 scores, and only two of them were of high quality.

Effect of DEX on incidence of nausea
Fourteen studies (three of them [15,16,17]   The incidence of vomiting was reported in fourteen studies (three of them [15,16,17] had subgroups),including 614 cases in the observation group and 610 cases in the control group. The heterogeneity test showed that I 2 = 29%, thus the xed-effects model was used. Meta-analysis results showed that the incidence of adverse reactions to vomiting in the observation group was lower than that in the control group, and the difference was statistically signi cant [OR = 0.12, 95%CI (0.09,0.18), P < 0.00001]. Fourteen studies (three of them [15,16,17] had subgroups), with 609 cases in observation group and 605 cases in control group, reported the incidence of facial ushing. Heterogeneity test showed that I 2 = 32% and xed effect model was selected for analysis. Results were shown in gure 4,which indicated that the incidence of  Fourteen studies (three of them [15,16,17] had subgroups), 609 cases in the observation group and 605 cases in the control group, reported the incidence of chest distress. Heterogeneity was detected and I 2 = 52%, thus random-effects model was selected for analysis. Results was shown in gure 5,which suggested that the incidence of adverse reactions to chest distress in the observation group was lower than that in the control group, and the difference was statistically signi cant [OR = 0.19, 95%CI(0.12,0.30), P < 0.00001].

Figure 5. Effect of DEX on incidence of chest distress
Effect of DEX on incidence of hypertension The incidence of hypertension was reported in nine of the included trials(one of them [16] included subgroups),318 cases in the observation group and 314 cases in the control group. Heterogeneity test showed I 2 = 36% and xed-effects model was selected. Results of meta-analysis were shown in gure 6. It indicated that the incidence of adverse reactions of hypertension in the observation group was lower than that in the control group signi cantly [OR = 0.09, 95%CI (0.05,0.14), P < 0.00001]. cases in the control group. According to heterogeneity test, I 2 = 30%, thus xed-effects model was applied.
Results was shown in gure 7, which showed that the incidence of adverse reactions to tachycardia in the observation group was signi cantly lower than that in the control group [OR = 0.04, 95%CI(0.02,0.07), P < 0.00001].

Discussion
Uterotonic agents such as oxytocin, carboprost tromethamine, carbetocin were commonly used drug in cesarean section for the prevention of postpartum hemorrhage. But except of uterine smooth muscle, these agents usually have effects on gastrointestinal tract, bronchus and vascular smooth muscle, due to the lack of smooth muscle selectivity, leading to nausea, vomiting, increased heart rate and other adverse reactions [23] .These side effects not only bring discomfort to the mothers, but also greatly increase the risk of perioperative surgery.
Dexmedetomidine, as a new selective alpha-2 adrenergic agonist, has a high selectivity to alpha-2 adrenergic receptor and has analgesic, sedative and anti-anxiety effects. It can block sympathetic during perioperative period, stabilize the cardiovascular system and reduce the need for anesthetics without affecting respiration [24] . The pharmacological mechanism of dexmedetomidine is multifaceted and complicated. On one hand, it acts on alpha-2 adrenalin receptors in presynaptic membrane and postsynaptic membrane of the spinal cord, inhibiting the release of adrenalin and breaking up the transmission of pain signals [25] .On the other hand, by activating postsynaptic membrane receptors, the concentration of catecholamines in blood can be reduced to alleviate the abnormal increase of blood pressure and heart rate caused by stress response [26] . Dexmedetomidine can also relieve spasm of gastrointestinal smooth muscle and blood vessels, and reduce the incidence of nausea, vomiting, chest and abdominal pain, facial ushing and other adverse reactions by anti-sympathetic effect and diastole of pulmonary and peripheral small blood vessels [27] .In addition, studies have shown that dexmedetomidine may reduce nausea and vomiting by directly acting on the locus coeruleus and inhibiting the vomiting center in the central nervous system that controls the downstream region of the medulla oblongae [28] .
Dexmedetomidine has variable pharmacological effects, but does it have an effect on adverse reactions caused by uterotonics, such as nausea, vomiting, facial ushing, chest distress, hypertension and tachycardias? To make a clear of it, we carry out a meta-analysis in this study. Our study evaluated the preventive effects of dexmedetomidine on the side effects of uterotonics in cesarean section surgery.
Eventually we included fteen studies with a total of 1174 patients. According to the results of meta-analysis, incidence of adverse reactions such as nausea, vomiting, facial ushing, chest distress, hypertension and tachycardia in DEX group were all lower than that in the control group, and the difference is remarkablely signi cant (P < 0.01), suggesting that dexmedetomidine can actually prevent and reduce the incidence of adverse reactions of uterotonics including nausea, vomiting, facial blushing, chest distress, hypertension and tachycardia during caesarean process.
Finally, our study has some limitations. Although all the literatures included in this study were RCTs, the number of articles and the sample size were still small. As for that most of the studies were from China and the quality of some literatures was not so high, there was inevitably the possibility of bias in the results. Most of the included literature did not report the total incidence of adverse reactions, so we did not conduct a bias analysis of the total incidence of adverse reactions in this study. In addition, there were differences in the dosage of dexmedetomidine and the way of administration of uterotonics in the included studies, which were not further evaluated in this paper due to the complicated classi cations.

Conclusions
In a word, we believe that in order to obtain reliable evidence of higher quality and provide more references for clinical medication, more high-quality, multi-center and large-sample of RCTs should be carried out in the future. In conclusion, the current meta-analysis suggests that dexmedetomidine is effective for the prevention of side-effects caused by uterotonics in caesarean section.

Declarations
Ethics approval and consent to participate Not applicable. Figure 1 Flow diagram of the study selection process Effect of DEX on incidence of facial ushing  Effect of DEX on incidence of hypertension Effect of DEX on incidence of tachycardia

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. PRISMA2009checklist.doc