Cesarean section deliveries (CSs) are the most frequent performed major operation in obstetrics [1]. Although CS can be life-saving for the mother and/or baby when used judiciously, it also carries risks for higher rates of maternal morbidity, mortality, and delayed recovery from the birth, difficulty establishing breastfeeding, neonatal morbidity, and admission to nursery care, in addition to the increase costs for the health system [2]. Pregnancies following CS also have increased risks for other types of abnormal placentation, reduced fetal growth, preterm birth, and possibly stillbirth [3]. Chronic maternal morbidities associated with CS include pelvic pain and adhesions, also adverse reproductive effects which may include decreased fertility and increased risk of spontaneous abortion and ectopic pregnancy. The WHO has considered a population-based rate of CS between 10-15%, as an ideal rate that was associated with a notable decline in maternal mortality ratio and neonatal mortality rate [4]. However the rate of CS has been increasing both in developed and developing countries in the past decade [1]. In this study, the purpose is to explore the different indication and trends that increase the CS rate in a teaching tertiary hospital comparing 1997 and 1998 (group A) and 2015 and 2016 (group B).

A retrospective study was conducted based on data collected from delivery‘s register and medical records for all CS in 1997, 1998 (group A) and 2015, 2016 (group B) at Jordan University Hospital. Demographic data, CS’s indications, and medical history were collected.

The indications for CS’s were based on the primary indication as stated by the attending obstetrician. Each delivery was assigned to the primary indication noted for that pregnancy, regardless of other reported indications. All CSs were allocated to one of main categories:, previous ≥ two CS, previous one CS, non reassuring fetal heart pattern (NFHRP), breech presentation and other malpresentation, labor dystocia, twin gestation and other multiple gestations, placentas previa, and accreta, maternal request, advanced maternal age (AMA), failed induction of labor (IOL), infertility and in vitro-fertilization (IVF), suspected macrosomia, small for gestational age (SGA), recurrent miscarriages, placental abruption, cephalopelvic disproportion (CPD), failed instrumental delivery, congenital fetal anomalies, chorioamnionitis, meconium stained liquor, and “others”.

The category of labor dystocia includes all types of obstructed or lack of progressive cervical dilation or lack of descent of the fetal head, or both. NRFHRP is defined as severe variable decelerations, late decelerations, prolonged decelerations (three to ten minutes), baseline bradycardia of < 100 beats/minute or base line tachycardia > 160 beats/minute. Other multiple gestations include triplets and quadruplets. The category malpresentation includes all types of malpresentations except cord and breech presentations, which were regarded as a separate category since 2001; a policy of planned CS for term breech presentation was introduced in the present department in accordance with recommendations of the Term Breech Trial Collaborative Group study [5]. Cord prolapsed and cord presentation were a separate group. AMA is when the mother is ≥ 35 year of age. Suspected macrosomia is when the estimated fetal weight (EFW) before delivery is ≥ 4,000 grams and SGA when the EFW is less than the 10% for the gestational age. Recurrent miscarriage is when the mother has ≥ two recurrent miscarriages and is on low molecular weight heparin in the indexed pregnancy. Infertility is when the women has primary or secondary infertility and had spontaneous pregnancy or got pregnant by induction of ovulation without IVF. All women with prelabor diagnosis of placenta previa regardless the type, were scheduled for a CS. All women with ≥ two CS were scheduled for elective surgery. Pregnancy induced hypertension (PIH) is when the mother blood pressure ≥ 140/90 at the time for decision of CS.

The study was approved by the Ethics Committee for Medical Research at the Jordan University Hospital and the University of Jordan. Data were analyzed using SPSS 20. Frequency and percentage were calculated for the categorical data and Pearson’s chisquared test. Fisher exact test was used when the cell is less than 5.0. Significance was set at p < 0.05.

In a total of 4,575 deliveries in group A and 5,895 in group B, 1,010 (22%) delivered by CS in group A compared to 1,907 (62%) and 1,835 (65%) in 2015 and 2016 in group B, respectively.

The leading indications for the increase rate of CS’s across the years were previous ≥ two CS. It increased from 5% (226/4,575) to 20 % (1,187/5,895), followed by decreasing rate of trial of previous uterine scar, the rate increased from 1.7% (79/4,575) up to 14% (832/5,895), NRFHP was 5.5 % (251/4,575) which increased to 7.5% (441/5,895), breech presentation was 2.6% (118/4575) which increased to 5.8% (341/5,895), twin pregnancies CS’s increased from 1% (47/4,575) to 3.1% (184/5,895), and dystocia was diagnosed in 1.4% (62/4575) compared to 2.6% (152/5,895); all with (p = 0.000), and placenta previa increased from 1.1% (49/4,757) to 1.7% (99/5,895) (p = 0.009) (Table 1).

Other indications were noted to increase across the years. Maternal request was not found in group A, however, there were 0.68% (40/5,895) women in group B, AMA increased from 0.07% (3/4575) compared to 0.41% (24/5,895), both with p = 0.000. Failed IOL was diagnosed in 0.17% (8/4,575) and 0.5% (29/5,895), with (p = 0.007). Infertility and IVF were causes for primary CS in the present study; 0.04% women (2/4,757) who had IVF underwent CS in group A, however after the increased use of IVF techniques, the rate of CS increased to 0.8% (48/5,895) in group B with (p = 0.000). Also, the rate of CS due to infertility increased from 0.02% (one woman) to 0.24% (14 women) with (p = 0.004). Mothers with suspected macrosomia was another indication for the increase of primary CS; the rate increased from 0.04% (two women) to 0.4% (22 women). The rate for fetuses who were SGA increased from 0.04% (two women) to 0.3% (19 women), and the rate of CS in women with recurrent miscarriages increased from 0.22% (ten women) to 1.3% (75 women); all with p = 0.000 (Table 1).

Whether or not to perform a cesarean section is based primarily on the question of what is best for or may save the lives of the mother and child [6]. The rate of CS worldwide is 18.6% (1.4%-56.4%), however it varies across different countries, worldwide: in Africa it is 7.3% (1.4%- 51.8%), in Asia 19.2% (1.7%-47.5%), in Europe 25.0% (13.9%-38.1%), in Latin America and the Caribbean 40.5% (5.5%-55.6%), in Northern America 32.3% (27.1%-32.8%), and in Oceania 31.1% (6.2%-33.4%) [7].

In previous studies in Jordan, the CS rate increased consistently, from 8.5% in 1990 to 12.9% in 1997, to 17.8% in 2002 [8] and continues to increase from 18.2% in 2002 in another study to 30.3% in 2012, with the most common reason for CS being “absence of a clear indication” [9].

In the present study the overall CS rate increased from 22% (group A) to 62% and 65% (2015 and 2016) in group B which is high (Table 1).

The leading cause for the present increase CS rate was ≥ two CS, as CS is an obligatory indication after two previous CS. [10]. The rate for ≥ two CS increased from 5% (226 women) to 20% (1,187 women) with p = 0.000, which is alarming and requires more attention.

The second cause for the increase rate of CS in the present study was a previous CS. This incidence is steadily rising worldwide [11], as the attempts of vaginal birth after CS is decreasing because of the increased risk of uterine rupture, repeat cesarean section, and greater association with perinatal risk than with elective repeated cesarean [12]. The trial of vaginal birth after CS at our hospital decreased; therefore the rate of CS increased from 1.7% (79 women) to 14% (832 women) with p = 0.000. Therefore it is important to carefully scrutinize the indication of primary CSs [11].

It is stated that NRFHRP is one of the leading primary indication for CS worldwide, and is the third cause in the present study; unfortunately, fetal distress is often over-diagnosed [13]. The rate of CS due to NRFHRP increased from 5.5 % (251 women) to 7.5 % (441 women) with p = 0.000), indicated that if fetal scalp pH was performed and the results were within the normal range, a CS could be avoided [14].

Breech presentation was the fourth cause in the present list for the increase of the CS rate in this hospital. Since an elective safe CS for breech presentation is much easier to execute and it requires less expertise and experience, as stated in some studies [5, 15], therefore as a consequence, almost all breech presenting babies are being delivered by CS. The present breech CS rate increased dramatically from 2.6% (118 women) to 5.8% (341 women) with p = 0.000. Using another calculation method, the present vaginal breech deliveries was 25% from the total breech fetuses in group A which decreased to 2% in group B (Table 2).

The incidence of twin, triplet, and higher-order multifetal gestations, has increased dramatically in recent years. Rates of CSs for multiple gestations also have increased, largely due to the perception that this will improve neonatal outcomes [16]. Although the Twin Birth Study found that planned vaginal lower segment CS is not advantageous to the fetuses [17], and there is insufficient evidence to support the routine use of planned CS for term twin pregnancy with leading cephalic presentation [18]. In spite of this, the rate of twin CS increased from 1% (47 women) to 3.1% (184 women) in the hospital with p = 0.000. Using another calculation method, the present twin vaginal deliveries decreased from 59% (67/114) women to 7% (14/198) women (Table 3).

Dystocia is another leading cause for primary CS in the United States, as it is rarely diagnosed with certainty [19], and also in this study. Dystocia was diagnosed in 1.4% (62 women) in group A, compared to 2.6% (152 women) in group B with p = 0.000, which indicates the need to be more patient and to use active management of the first stage of labor, with strict diagnosis of onset of labor [20].

With time there are higher incidences of more cases of placenta previa and accreta, as it is known fact that the risk of placenta previa increases with the number of previous CSs [21] Hence the rate of placenta previa increased from 1.1% (49 women) to 1.7% (99) women with p = 0.009. There was no single case of placenta accreta in group A, however in Group B there were four cesarean hysterectomies due to placenta accreta, and if cesarean rates continue to increase, the annual incidence of placenta previa, placenta accreta, and maternal death will also substantially increase [22].

CS was high in women with recurrent miscarriage in one study [23] and this was noted here as a cause for the increase rates of primary CS. In group A 0.22% (ten women) had CS compared to 1.3% (75 women) in group B (p = 0.000).

Infertility and IVF are causes for primary CS and the rates appear to be disproportionately high in term singleton births [24] and multiple gestations [25]. In group A, only 0.04% (two women) who had IVF underwent CS; however after the increase use of IVF in our country, the rate of CS in group B increased to 0.8% (48 women), with p = 0.000.The rate of CS due to infertility also increased from 0.02% (one woman) in group A to 0.24% (14 women) in group B with (p = 0.004).

In consistent with other parts of the world, new indications were introduced as maternal request for CS without any medical indication [26]. Although it is known that this carries the potential risks for a longer maternal hospital stay, an increased risk of respiratory problems for the infant and greater complications in subsequent pregnancies [27]. In group B, 0.68% (40 women) had CS, where there were not any in group A.

Compared to spontaneous onset of delivery, IOL is associated with an increased risk of emergency CS both among nulliparous and multiparous women [28] The rate of CS due to failed IOL rose from 0.17% (eight women) to 0.5% (29 women), with p = 0.007. AMA is noted in some studies to increase the risk for CSs both for nulliparasand multiparas [29]. In spite of the increase incidence of comorbidities and pregnancy-related complications [30], there were three (0.07%) women who had CS in group A compared to 24 (0.41%) in group B with p = 0.000. Delivery of macrosomic fetuses is not always associated with perinatal complications, however attempts to eliminate these complications lead to an increase in the number of CS and labor inductions, in spite of the knowledge that clinical and ultrasonographic examination cannot exclude or confirm the possibility of macrosomia with sufficient specificity and sensitivity [31]. Mothers with suspected fetal macrosomia were scheduled for primary CS, therefore the rate increased from 0.04 % (two women) to 0.4% (22 women) with p = 0.0003.

In the absence of other obstetric indications requiring CS, there is no sufficient evidence to recommend planned CS in the pregnancy with SGA aiming to reduce neonatal and perinatal morbidity and mortality [32]. However the present CS rate increased from 0.04% (two women) to 0.3% (19 women) with p = 0.000.

Other indications for CSs were not significant in the present study throughout the years, as failed instrumental delivery, PIH, placental abruption, triplets and quadruplets, congenital fetal anomaly, malpresentations, CPD, chorioamniontis, and meconium stained liquor.

The present rate is high and could be partially explain ed by the fact that the pregnant women attending the hospital are mainly from the middle class, those who are employed, those that do not desire to have many children, those who had primary CS in the private sector for an unconvincing indication, and those do not want to undergo a vaginal delivery. Nevertheless, although the principle contributor is previous ≥ two CS, we feel that in order to lower the total CS rate; the appropriate way would be through reducing the primary section by increasing attempts of vaginal birth after CS, tightening the criteria for NRFHRP (and liberal use of scalp pH), encouraging external cephalic version for breech presentation, maintaining operative vaginal breech delivery skills, vaginal delivery for twin gestations when the first twin is in cephalic presentation, and applying more strict criteria for intervention for labor dystocia. However proper counseling for women who asked for CS without any medical reasons and for those who were above 35 years of age to alleviate their fears, proper management for multiparous women in labor, labor induction at complete 41 weeks’ gestation, judicious decisions for route of delivery in women with infertility, proper fetal weight estimation with careful review of the previous obstetrical history in women with suspected macrosomia, strict intrapartum fetal monitoring for SGA in labor, and proper management of women with recurrent miscarriage, will all assist to reduce the rate of primary CS and hence the overall rate of the CS. The study has few limitations, since it is a retrospective study, some confounders could not be excluded as a possible explanation for the findings.

CS rates have steadily and dramatically increased over the last 20 years of study. The appropriate way to deal with this global burden is to reduce the primary section. Clinicians and women need to be aware of the long-term risks associated with CS when determining the method of delivery for first and subsequent births.