A Secondary Analysis of the LILAC Study Regarding Cesarean Birth by Maternal Request in Women Living with HIV.

BACKGROUND: Latin America has the highest regional average cesarean birth rates. One potential driver is cesarean birth by maternal request (CBMR).
METHODS: We analyzed of a large prospective cohort study of HIV-infected women in six Latin American countries.
RESULTS: Comparisons were made between women who chose CBMR (n = 38) and women with a medical indication for cesarean (n = 683). The only variable associated with CBMR was onset of labor (AOR 0.3 [0.1,0.9], p = 0.04). CONSLUSION: Spontaneous labor reduced the likelihood of a woman living with HIV to pursue CBMR in a large Latin American cohort.

Introduction

Latin America has the highest regional average cesarean birth rates in the world [1]. One potential driver is cesarean birth by maternal request (CBMR), which is elective cesarean birth with no other medical indication; this practice is advised against by the World Health Organization [2]. The objective of this analysis was to contribute additional data to the literature on CBMR, an elusive practice in Latin America and the Caribbean as well as globally.

Methods

The International Site Development Initiative (NISDI) Perinatal Study (2002–2007) and subsequent Longitudinal Study in Latin American Countries (LILAC) (2008–2012) are prospective cohort studies of HIV-infected women in six countries (Argentina, Bahamas, Brazil, Jamaica, Mexico, and Peru) [3]. Detailed methods are published, but the objectives of the protocols were to describe care of women living with HIV and their neonates in pregnancy and the early postpartum period [3].

Results

Women who underwent cesarean for the indication of “patient request” or “patient desires sterilization” were considered to have received CBMR (5.3%) for the purposes of this analysis. All other indications were considered medically necessary: abruption [1.0%], genital infection [1.5%], anticipated cephalon-pelvic disproportion [0.1%], arrest disorder [2.2%], cephalon-pelvic disproportion [3.5%], cord prolapse [0.3%], eclampsia/pre-eclampsia [2.1%], failed induction [3.9%], malpresentation [2.5%], fetal heart rate [6.1%], oligohydramnios [1.8%], other [2.0%], placenta previa [0.6%], prevention of HIV infection [44.1%], prolonged rupture of membranes [3.9%], and elective repeat cesarean birth [19.1%]. Comparisons were made between women who chose CBMR (n = 38) and women with a medical indication for cesarean birth (n = 683); this comprised the population of 721 (59.8%) women out of 1206 in the cohort whom delivered by cesarean birth. The only variable associated with CBMR in this cohort of pregnant women living with HIV was onset of labor (AOR 0.3 [0.1,0.9], p = 0.04), suggesting that the spontaneous onset of labor reduced the likelihood of a woman living with HIV to pursue CBMR.

Discussion

Onset of labor can be modified by membrane stripping, if permitted in this subpopulation [4]. This study is limited by the fact that the data is old, that outcomes were not included, that the sample is weighted toward Brazil and Argentina, and that the datasets were not designed for the study question [3]. However, we believe documenting the prevalence and practice of CBMR in any population is a contribution to the literature and knowing that spontaneous labor contributes to a reduction in the practice in this particular population, is of value.

Table 1:

Bivariate comparisons and multivariable modeling of characteristics associated with cesarean birth by maternal request in a Latin American and Caribbean cohort of women living with human immunodeficiency virus.

Bivariate Comparisons
Characteristic	N (%) N = 721	Medically Necessary (n =683, 94.7%)	CBMR (n =38, 5.3%)	P-Value
Age in years, Median (IQR)	28 [24, 33]	27 [24, 33]	30 [25, 33]	0.10[a]
Years of Education, Median (IQR)	8 [6, 11]	8 [6, 10]	9 [7, 11]	0.22[a]
Relationship Status				1.0[b]
Single	188 (26.1%)	178 (26.1%)	10 (26.3%)
Not single	533 (73.9%)	505 (73.9%)	28 (73.7%)
Employed				0.05[b]
Student	35 (4.9%)	30 (4.4%)	5 (13.2%)
No	11 (1.5%)	11 (1.6%)	0 (0.0%)
Yes	675 (93.6%)	642 (94.0%)	33 (86.8%)
Parity, Median (IQR)	1 [1, 3]	1 [1, 3]	1 [1, 2]	0.59[a]
Language				0.009[b]
English	17 (2.4%)	17 (2.5%)	0 (0.0%)
Haitian Creole	3 (0.4%)	3 (0.4%)	0 (0.0%)
Portuguese	396 (54.9%)	365 (53.4%)	31 (81.6%)
Spanish	305 (42.3%)	298 (43.6%)	7 (18.4%)
Country				0.05[b]
Argentina	238 (33.0%)	231 (33.8%)	7 (18.4%)
Bahamas	13 (1.8%)	13 (1.9%)	0 (0.0%)
Brazil	397 (55.0%)	366 (53.6%)	31 (81.6%)
Jamaica	7 (1.0%)	7 (1.0%)	0 (0.0%)
Mexico	40 (5.6%)	40 (5.9%)	0 (0.0%)
Peru	26 (3.6%)	26 (3.8%)	0 (0.0%)
Intended Delivery Mode				0.60[b]
Cesarean Birth	326 (45.2%)	308 (45.1%)	18 (47.4%)
Vaginal Birth	291 (40.4%)	278 (40.7%)	13 (34.2%)
Unsure	104 (14.4%)	97 (14.2%)	7 (18.4%)
Delivery Location				0.11[b]
Hospital	686 (95.2%)	652 (95.5%)	34 (89.5%)
Other non-study Medical Facility	35 (4.8%)	31 (4.5%)	4 (10.5%)
Spontaneous Onset of Labor				0.07[b]
No	530 (73.5%)	496 (72.6%)	34 (89.5%)
Yes	186 (25.8%)	182 (26.7%)	4 (10.5%)
Unknown	5 (0.7%)	5 (0.7%)	0 (0.0%)
Gestational Age, Median (IQR)	38.3 [37.3, 39.1]	38.3 [37.3, 39.1]	38.3 [37.7, 38.6]	0.68[a]
Multivariable Model[c]
	OR	CI	P-value
Age (continuous)	1.1	1.0,11	0.07
Employed (ref: unemployed, including student)	0.4	0.1,11	0.07
Speaks Spanish (ref: speaks another language)	3.8	0.1,103.1	0.4
Brazilian (ref: any other nationality)	13.3	0.5,359.5	0.1
Delivered in Hospital (ref: other non-study medical facility)	0.4	0.1,13	0.1
Spontaneous Labor (ref: non-spontaneous labor)	0.3	0.1,0.9	0.04

Kruskall-Wallis test

Fisher’s Exact test

Logistic Regression

CBMR: cesarean birth by maternal request