General versus Neuraxial Anesthesia in Cesarean Section : A Systematic Review

This is an open-access article, published by Evidence Based Communications (EBC). This work is licensed under the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium or format for any lawful purpose.To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. From Department of Anesthesiology, West China Hospital, Sichuan University, Chengdu, China; Department of Anesthesiology, Sichuan Provincial People's Hospital, Chengdu, China.


C
esarean section is a procedure of baby delivery through an incision of abdominal and uterus wall of mothers.It is an alternative way if vaginal delivery is possible to cause harm to mothers or babies.
Both of neuraxial and general anesthesia techniques are considered in cesarean section.Neuraxial anesthesia technique included spinal, epidural or combined spinal and epidural anesthesia.Due to the minimal anesthetic exposure to babies and possible risk of general anesthesia such as difficult airway, neuraxial techniques are preferred anesthesia methods for cesarean section (1).However, general anesthesia is a faster technique than neuraxial anesthesia.So it would be a prior anesthesia technique in some emergent situation.Besides, neuraxial anesthesia was contraindicated in some situation, such as coagulation disorders.Therefore, general anesthesia was still an important technique in cesarean section.
As more comprehensive understanding of pregnant physiology and more use of short-acting anesthetic agents in recent years, the safety of general anesthesia improved well (2).It is important to clarify the safety of general anesthesia and neuraxial anesthesia for cesarean section.

Types of Studies
Only randomized controlled trials (RCTs) which compared general anesthesia with neuraxial anesthesia including spinal anesthesia, epidural anesthesia or combined spinal and epidural anesthesia in cesarean section were included.We restricted the included articles in English or Chinese.

Types of Participants
Mothers with or without any complication underwent cesarean section were included.

Types of Interventions
General anesthesia constituted to intervention group.Neuraxial anesthesia including spinal, epidural or combined spinal and epidural anesthesia constituted to control group.We compared the intervention group versus control group.

Types of Outcome Measures
Primary outcomes included neonatal mortality, risk of neonatal Apgar score < 7 at 1 and 5 minutes after delivery.Secondary outcomes included mean neonatal Apgar scores at 1 and 5 minutes after delivery, need for oxygen by mask or intubation in newborns, risk for receiving postoperative blood transfusion in mothers, mean maternal blood loss, maternal adverse events such as risk of postoperative nausea and vomiting, headache, pruritus and shivering.

Study Selection
Two authors independently screened all records for eligibility based on their titles, abstracts and keywords.We retrieved the full published version of the selected studies for further assessment.Only those studies completely met the inclusion criteria were included.We resolved any disagreements by discussing and consulting with a third author.

Data Collection
Two authors independently extracted data from included studies.The extracted information included age, complications, anesthesia drugs, primary and secondary outcomes.We resolved any disagreement by discussion.

Risk of Bias in Included Studies
We performed risk of bias assessment by using 'Risk of bias' tool described in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions 3. We assessed the 7 domains as follows for each study: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and any other potential bias.The judgment for each domain was low risk, high risk or unclear.
We considered a trial as having low risk of bias if all domains were assessed as low risk.We considered a trial as having high risk of bias if one or more domains were assessed as high or unclear risk.

Summary Measures
We calculated the risk ratio (RR) with 95% confidence intervals (CIs) for the dichotomous data.We also used the mean difference (MD) with 95% CIs for continuous data if same scales were used in all studies.

Synthesis of Results
We performed all analysis by using Review Manager 5.3.We used random-effect model as the heterogeneity exist between trials.
We estimated mean and SD from the median and range as described by Hozo (4).We assessed the statistical heterogeneity according to both Chi 2 test and I 2 statistic.For the Chi 2 test, we considered the statistical significance with P value <0.1.I 2 statistic indicated the percentage of heterogeneity impact on the meta-analysis.We considered heterogeneity to be substantial if the I 2 >50%.

Sensitive Analysis
We performed sensitive analysis as excluding the studies with high risk of bias to determine the effect of methodological quality on the results.

Studies Characteristics
Included studies characteristics were shown in Table.

Risk of Bias in Included Studies
Risk of bias summary and graph were shown in Figure 2 and Figure 3.Only 3 studies were judged as low risk of bias (10,15,18).Selection bias and attrition bias were the main problems in methodology.

Sequence Generation (Selection Bias)
We judged this item as low risk of bias in 7 studies, five of which used computer-generated random number table (5-7, 10, 13) and two of which generated sequence by drawing lots (15,18).We judged this item as unclear risk of bias in the rest eight included studies, because they did not describe detailed randomized method (8,9,11,12,14,16,17,19).

Allocation Concealment (Selection Bias)
We judged this item as low risk of bias in 4 studies, two of which concealed allocation in sealed envelopes (7,10) and two of which by drawing lots (15,18).We judged this item as high risk of

Blinding (Performance Bias and Detection Bias)
No studies mentioned blind they used, but the primary outcomes were not influenced by blind method.So we judged low risk of bias for all included studies.

Incomplete Outcome Data (Attrition Bias)
Missing data occupied more than 15% in 2 studies (7,11).As the primary outcomes were low incidence events, so we judged 2 studies as high risk of bias.The other included studies were judged as low risk of bias.

Selective Reporting (Reporting Bias) and Other Potential Sources of Bias
None selective reporting or other potential bias were detected in all included studies.

Neonatal Mortality
Only one study (6) reported neonatal mortality.It reported that no neonatal death occurred in general anesthesia or neuraxial anesthesia group.

Risk of neonatal Apgar score <7 at 1 and 5 minutes after delivery
Five studies (10,13,15,17,18) reported risk of Apgar score <7 at 1 minute after delivery.It indicated that newborns had higher risk of lower than 7 Apgar score at 1 minute after delivery in maternal general anesthesia group (RR 4.81, 95% CI 1.72 to 13.46) (Figure 4).But two studies (17,18) reported that the risk of Apgar  5).

Risk for Oxygen by Mask or Intubation in Newborns
Two studies (10,11) reported risk of oxygen by mask or intubation in newborns.The result indicated that no significant difference between maternal general anesthesia group and neuraxial anesthesia group (RR 1.23, 95% CI 0.33 to 4.53).

Mean Maternal Blood Loss
Two studies (6, 9) reported maternal blood loss.It indicated that there were more blood loss in maternal general anesthesia group than neuraxial anesthesia group (MD 75.8, 95% CI 21.18 to 130.41).

Risk for Receiving Postoperative Blood Transfusion in Mothers
Three studies (5,6,9) reported the risk for receiving postoperative blood transfusion in mothers.The result showed that there was no significant difference between maternal general anesthesia group and neuraxial anesthesia group (RR 2.85, 95% CI 0.93 to 8.72).

Risk of Maternal Ddverse Events
Five studies (6, 9, 10, 14, 16) reported risk of postoperative nausea, with no significant difference between maternal general anesthesia group and neuraxial anesthesia group (RR 1.47, 95% CI 0.99 to 2.17).Four studies (6, 10, 14, 16) reported risk of postoperative vomiting.The result demonstrated that there was significantly higher risk in maternal general anesthesia group than neuraxial anesthesia group (RR 4.13, 95% CI 1.41 to 12.09).One study (16) reported risk of shivering, and there was significantly higher risk in maternal general anesthesia group than neuraxial anesthesia group (RR 8.00, 95% CI 1.14 to 56.33).Two studies (6,13) reported risk of headache.The result indicated that there was significantly lower risk in maternal general anesthesia group than neuraxial anesthesia group (RR 0.11, 95% CI 0.01 to 0.87).Two studies (9,14) reported risk of pruritus, with significantly lower risk in maternal general anesthesia group than neuraxial anesthesia group (RR 0.13, 95% CI 0.02 to 0.99).

Sensitive Analysis and Subgroup Analysis
We performed sensitive analysis as excluding studies with high risk of bias.Therefore, only three studies (10,15,18) were included.The result of mean neonatal Apgar score at 5 minutes after delivery was changed (MD -0.26, 95% CI -0.80 to 0.28).The rest results as mean neonatal Apgar score at 1 minutes after delivery, risk of neonatal Apgar score < 7 at 1 and 5 minutes after delivery, risk for oxygen by mask or intubation in newborns and risk of postoperative nausea was not changed.Heterogeinity was still significant when we performed subgroup analysis according to different neuraxial anesthesia technique, with or without maternal complications or whether using muscle relaxant before delivery or not.included studies.In these included studies, 11 of them (5, 6, 8, 10-15, 17, 18) with 1242 mothers compared general anesthesia to spinal anesthe-sia, 3 of them (9,16,19) with 117 mothers compared general anesthesia to epidural anesthesia and 1 of them with 35 mothers compared gener-    al anesthesia to combined spinal and epidural anesthesia.

Summary of Evidence
For neonatal outcomes, at 1 minute after delivery, we found Apgar score was significantly lower in maternal general anesthesia group.Meanwhile, the risk of newborns for Apgar score lower than 7 was higher when mother underwent general anesthesia.These results suggested that anesthesia technique would affect the newborn's immediate condition after delivery.The risk of hypoxia may be higher in maternal general anesthesia group.But for the later condition, at 5 minutes after delivery, the risk of newborns for Apgar score lower than 7 was similar under general or neuraxial anesthesia.For the non-surrogate outcomes, such as neonatal mortality and risk of oxygen by mask or intubation, the results were similar in different anesthesia technique.The findings indicated that general anesthesia was a safe alternative technique for newborns, but the risk of early hypoxia should be paid attention.
For maternal outcomes, the mean blood loss was more in general anesthesia group, but the risk for receiving postoperative blood transfusion in mothers was similar in general and neuraxial anesthesia groups.It indicated that general anesthesia was possibly related to more maternal bleeding during cesarean section, although the 2 included studies had high risk of selection bias.Besides, the risk of shivering, nausea and vomiting was higher in general anesthesia group.But the risk of headache and pruritus was higher in neuraxial anesthesia group.Recently, The U.S. Food and Drug Administration (FDA) released a warning of general anesthetics on pregnant woman (20).It warned that more than 3 hours use of general anesthetics in pregnant woman during their third trimester may affect the neuron development in newborn babies.So even the above adverse events we listed were not fatal, we should also concerned seriously the necessity of general anesthesia in cesarean section.Practically, the majority of cesarean section was no more than 3 hours.Therefore, if there were some contraindications of neuraxial anesthesia for mothers, such as bleeding tendency, or emergent condition, general anesthesia should be considered as an alternative anesthesia technique.
Due to including articles published in recent 5 years, we found mean Apgar score was lower at 1 and 5 minutes after delivery in maternal general anesthesia group, which was different from other systematic review before (21).Thus, even there was no death or other severe complications found in maternal general anesthesia group, we still recommended neuraxial anesthesia as first choice in cesarean section.In fact, it was consistent with our clinical practice.A clinical practice survey in Europe, spinal anesthesia was the most favored technique among anesthesiologists, and general anesthesia was an alternative technique in some particular mothers (22).

Limitations
Firstly, according to the methodological assessment, only three of included studies were considered as low risk of bias.Although the three studies included 300 mothers and the results was not changed in sensitive analysis, we also considered well-designed large sample sized RCTs were needed.What's more, the restriction of language could either miss some potential useful data.

CONCLUSIONS
For clinical practice, we recommended neuraxial anesthesia as first choice in cesarean section, because more adverse effect were reported both in newborns and mothers under general anesthesia.
For further clinical research, more non-surrogate outcomes should be reported, such as maternal and neonatal mortality and morbidity.
A total of 1394 mothers were involved in the 15

Figure 6 .
Figure 6.Mean Neonatal Apgar Scores at 1 Minute after Delivery.

Figure 7 .
Figure 7. Mean Neonatal Apgar Scores at 5 Minutes after Delivery.