Previous published data classified cesarean, which were not performed as an elective cesarean, in an inconsistent way. The cases, classified as emergency cesareans, were very inhomogeneous. Lucas et al. performed a study in the year 2000 which had the aim to set a new classification for emergency cesareans, and they proposed the following:

1) Emergency cesarean: immediate threat of life for woman or fetus;

2) Urgent: maternal or fetal compromise which is not immediately life threatening;

3) Scheduled: needing early delivery but no maternal or fetal compromise;

4) Elective: at a time to suit the woman and the maternity team [1] .

Emergency cesarean section (ECS) is conducted, when a situation requires immediate action for decreasing the risk to mother and/or child, e.g. placental abruption, bleeding caused by placenta previa or imminent intrauterine asphyxia [2] [3] . An urgent cesarean section (UCS) was conducted during birth process, though a vaginal birth was planned, e.g. a protracted birth [3] . The performance and specifically the time management of emergency cesarean section were discussed in many studies, especially regarding the decision-to-delivery interval (DDI).

The DDI is the time interval between the decision to perform a cesarean and the delivery of the child [4] [5] . The German society for gynecology and obstetrics allows a DDI of 20 minutes, while the American College of Obstetricians and Gynecologists and the Royal College of Obstetricians and Gynecologists recommend a DDI of 30 minutes [6] [7] .

To obey the recommendation of 20 minutes in our institution, a tertiary care hospital with 3500 births per year, the management of emergency cesareans was adapted to the following:

1) On admission the patients are screened for risk factors which could predict complications, e.g. twin pregnancies, former cesareans, especially former emergency of urgent cesarean sections or intrauterine fetal retardation. If risk factors are found, the patients are admitted in a special equipped delivery room (with equipment as lamps or anesthesia machine for an emergency cesarean).

2) If these patients would need an emergency cesarean, it is directly performed in the delivery room.

3) After consultation with the anesthesiologist women without former epidural anesthesia received general anesthesia.

4) The obstetric staff is equipped with an ECS beeper. When ECS alarm is given, the staff has to call back immediately. In the ECS protocol the time of the alarm, the call back time, the indication for the ECS, the gestational age, the birth weight, the pH and the Apgar score are registered.

Due to consequent implementation of these steps and good interdisciplinary communication, the DDI decreased from average 15.39 to 9.43 minutes.

Although the recommended 20 minutes are an accurate goal in most patients, the situation itself has high stress potential for women and child, and the life threatening situation leading to the decision to undertake an ECS could also affect the outcome. Recommendations for time management for UCS do not exist.

The aim of the study was to investigate the maternal and neonatal outcomes for ECS and UCS, in respect.

The study was conducted as a case-control study. In the study period from 2003 to 2012 all women undergoing a cesarean section in our institution were recorded in an in-house database. All women undergoing an ECS in the study period (singleton and twin pregnancies) were included in the case group. Controls were matched by age.

All children (singleton and twins) born by ECS during the study period were included in the case group, controls were matched by sex and gestational age.

In the case group 303 women and 336 children were included and 303 women and 336 children were matched as controls.

The maternal outcome was defined by blood loss, hemorrhage, integrity of the placenta, anomaly of placental separation, eclampsia, hysterectomy, puerperal fever, sepsis, anemia and death. The new-born outcome was defined by stillbirth, Apgar score, umbilical cord pH, admission to NICU and death within seven days.

The study was approved by the local ethic committee (#1282/2013, 17.04.2013).

First descriptive analyses were performed to describe both groups. Epidemiologic and anamnestic parameters were compared, e.g. age, obstetric parameters, gestational age, weight of the new-born, sex or malformations. The frequency of maternal and neonatal morbidities were investigated and compared between cases and controls, also regarding singleton and twin pregnancies. The Wilcoxon test and the Chi-square-test were conducted. p < 0.05 was considered statistically significant. All statistical analyses were performed with IBM SPSS Statistics 19 for windows.

An unpredictable event, which leads to the conduction of an ECS, is a strong stressor for the mother and her child. The additional stress may influence the maternal and neonatal outcome. Differences in the maternal and

neonatal outcome are partially explainable regarding the underlying pathologies. Indications for ECS are acute life threatening events, whereas UCS are conducted for maternal or fetal compromise which is not immediately life threatening.

Studies have shown that a shorter DDI is associated with poorer fetal outcomes [8] -[10] , tough the hypoxic event is finished earlier. Theories to explain this phenomenon include, that the anxiety for the woman generated by the decision for ECS may provoke increased maternal catecholamine release that initially causes reduced perfusion of the placental bed and gas exchange, causing temporal fetal acidosis [8] . Even though the uterine vessels should be constricted due to catecholamine release, the blood loss of women undergoing ECS was higher than in women undergoing UCS. The vasoconstriction of blood vessels in the uterus is induced by norepinephrine, whereas epinephrine causes vasodilatation. Segal et al. showed that the beginning of efficient analgesia reduces the level of epinephrine [11] . The analgesia at ECS is probably not as efficient as at UCS because of the fast induction of anesthesia. However not every circumstance having influence to the blood loss is known.

Women undergoing ECS were in the 36^th week of gestation, whereas women with UCS were in the 38^th week. This fact surely affects the outcome, especially the neonatal outcome. The mortality of “late preterm” is higher compared with mature new-born [12] . Even though in this study the neonates were matched by gestational week to diminish this influence of gestational age at the mortality, the perinatal mortality was higher for ECS than UCS (8.3% vs. 1.7%). The perinatal mortality was also influenced by the fact, if the pregnancy was singleton or multiple. Singletons had a perinatal mortality of 3.8%, twins of 8.1%. Singletons born with ECS had a higher mortality than singletons born with UCS (5.7% vs. 1.4%). Twins born with ECS had a higher mortality than twins born with UCS (17.41% vs. 2.63%). Admission to NICU was more frequent for twins; in the case group the probability was higher than in the control group. (ECS vs. UCS singletons 54.26% vs. 35.22%, twins 77.7% vs. 63.72%). Other studies showed that the perinatal mortality is higher for twins than for singletons and that twins were more often transferred to NICU [13] [14] . The Apgar score after one, five and ten minutes was different in the case and the control groups. The Apgar score was initially conceived as predictor for survival and to evaluate resuscitation, but not to predict the long term outcome. The score is not only influenced by gestational age, maternal medication and resuscitation but also by neurological and respiratory complications. The five-minute Apgar score is not a good marker for an acute hypoxic event during birth [15] [16] . In addition the condition leading to the conduction of an ECS may have influence on the Apgar levels. In our study singletons had a higher five and ten minute Apgar score than twins. Although the Apgar levels were significantly lower in the case group, the validity is questionable because most of the children in the case and the control groups achieved high levels from 7 - 10, which are classified as normal [15] .

The pH values were lower in the case group (median 7.20) than in the control group (median 7.29). Eskes et al. had shown that at fetal bradycardia, which is often associated with hypoxia, the pH value declines every minute about 0.006 units [17] . This may also have influence on the lower pH in the case group. Interestingly the umbilical cord pH value was lower for singletons than for twins, especially in the case group. Georgieva et al. had shown that the umbilical cord pH is a good predictor for low Apgar score, need for resuscitation, ventilation with mask, seizures and other cerebral problems including death [18] .

Various studies had shown that the second born twin has a poorer outcome than the first born [19] [20] . This could not be repeated in this study.

In summary, the maternal and fetal outcome is poorer in ECS than UCS. The maternal blood loss is higher, hemorrhage is more frequent, just as puerperal anemia. Apgar levels and umbilical cord pH values are lower at neonates born by ECS. Twins had lower Apgar levels, but higher umbilical cord pH values. The perinatal mortality is higher for singletons than for twins.

The study has some limitations, in parts caused by the retrospective design with possible errors in data selection. The number of investigated patients was low, especially for twin pregnancies, which could also reduce the validity of the study.

Comparison of the maternal and neonatal outcome for singletons and twins between emergency and urgent cesarean sections.

The authors report no conflict of interest.