Oral Contraceptives and Endometriosis

The Past Use of Oral Contraceptives for Treating Severe Primary Dysmenorrhea Is Associated With Endometriosis, Especially Deep Infiltrating Endometriosis

Charles Chapron; Carlos Souza; Bruno Borghese; Marie-Christine Lafay-Pillet; Pietro Santulli; Gérard Bijaoui; François Goffinet; Dominique de Ziegler

Disclosures

Hum Reprod. 2011;26(8):2028-2035.

Abstract and Introduction

Abstract

Background: The relationship between the use of oral contraception (OC) and endometriosis remains controversial. We therefore compared various characteristics of OC use and the surgical diagnosis of endometriosis histologically graded as superficial peritoneal endometriosis (SUP), ovarian endometrioma (OMA) or deep infiltrating endometriosis (DIE).
Methods: This cross-sectional study included 566 patients without visible endometriosis at surgery as controls, and 410 patients with histologically proven endometriosis, categorized by their worst lesions as SUP n = 47, OMA n = 120 and DIE n = 243. Personal data, including on OC use, were prospectively collected during standardized interviews. Statistical analysis was performed using unconditional logistic regression.
Results: Past OC users had an increased incidence of endometriosis (adjusted odd ratios (OR) = 2.79, 95% confidence interval (CI) 1.74–5.12, P = 0.002) of any revised American Fertility Society stage. Women who had previously used OC for severe primary dysmenorrhea were even more frequently diagnosed with endometriosis (adjusted OR = 5.6, 95% CI 3.2–9.8), especially for DIE (adjusted OR = 16.2, 95% CI 7.8–35.3). Women who had previously used OC for other reasons also had an increased risk of endometriosis, but to a lesser extent (adjusted OR = 2.6, 95% CI 1.8–4.1). The age at which OC was initiated, duration of OC use and free interval from last OC use were not significantly different between control and endometriosis women, irrespective of histological grading. Current OC users did not show an increased prevalence of endometriosis (OR = 1.22, 95% CI 0.6–2.52).
Conclusions: Our data indicate that a history of OC use for severe primary dysmenorrhea is associated with surgical diagnosis of endometriosis, especially DIE, later in life. However, this does not necessarily mean that use of OC increases the risk of developing endometriosis. Past use of OC for primary dysmenorrhea may serve as a marker for women with endometriosis and DIE.

Introduction

Endometriosis, histologically defined as functional endometrial glands and stroma developing outside of the uterine cavity (Sampson, 1927), is an enigmatic disease (Borghese et al., 2008; Bulun, 2009; Ngo et al., 2009) responsible for pelvic pain (Fauconnier and Chapron, 2005) and infertility (Matzuk and Lamb, 2008; de Ziegler et al., 2010). Public health wise, endometriosis is the source of an important economic burden (Gao et al., 2006). A common early clinical symptom of endometriosis is an enduring history of dysmenorrhea, which is often primary and severe (Momoeda et al., 2002).

Oral contraception (OC) is the treatment commonly offered to young women suffering from dysmenorrhea that is not adequately alleviated by non-steroidal anti-inflammatory drugs (NSAIDs) (Milsom et al., 1990; Harel, 2008). Suppressing ovarian function with OC improves the symptoms of dysmenorrhea (Group, 2005), but these classically recur upon stopping (Harada et al., 2008). Practically therefore, OC is often continued for several years (ACOG, 2005), as it has been shown to effectively prevent recurrence (Vercellini et al., 2008, 2010).

Not unexpectedly, the possibility of a link between OC use and endometriosis is a debated one (Hemmings et al., 2004; Missmer et al., 2004; Vercellini et al., 2011). In a case–control study, Parazzini et al., (1994) reported that using OC increases the risk of endometriosis. Conversely, other researchers observed, in a large cohort study, a lower risk of endometriosis amongst OC users (Vessey et al., 1993). Finally, still other studies failed to find any association between use of OC pills and endometriosis (Darrow et al., 1993; Heilier et al., 2007). We see two possible causes for these contradictory findings: (i) early reports (Darrow et al., 1993; Vessey et al., 1993; Calhaz-Jorge et al., 2004; Heilier et al., 2007) failed to mention for which indication OC was prescribed (Somigliana et al., 2011): treating severe primary dysmenorrhea, bleeding disorders, secondary dysmenorrhea and/or pelvic pain or simply for contraception; and (ii) in most prior studies, endometriosis was not histologically staged (Vessey et al., 1993; Parazzini et al., 1994; Hemmings et al., 2004).

Realizing that the link between endometriosis and OC use is still unclear, we undertook to reassess this issue while accounting for the two likely sources of confusion, the indication of OC and the histological staging of endometriosis. In the present study, we therefore determined whether OC has been used for treating severe primary dysmenorrhea or other reasons as well as other parameters pertinent to OC use. Moreover, endometriosis was histologically categorized as (i) superficial peritoneal endometriosis (SUP), (ii) ovarian endometrioma (OMA) or (iii) deep infiltrating endometriosis (DIE).

1 2 3 4

Next Section

Cite this: Oral Contraceptives and Endometriosis - Medscape - Aug 01, 2011.

Table I. Patients' characteristics between control and endometriosis groups.
Patients' characteristics	Group A (n = 566)	Group B (n = 410)	P
Age (years)^a	32.18 ± 5.82	32.08 ± 5.46	0.78^d
Gravidity (n, %)^a	1. 01 ± 1.42	0.61 ± 1.08	<0.0001^d
Parity (n, %)^a	0.5 ± 0.99	0.3 ± 0.67	<0.0001^d
Height (cm)^a	167.9 ± 65.4	165.19 ± 7.07	0.39^d
Weight (kg)^a	63.8 ± 12.05	60.1 ± 10.46	<0.0001^d
Body mass index (kg/m²)^a	23.33 ± 4.45	22.07 ± 3.92	<0.0001^d
Age at menarche (years)^a	12.92 ± 1.64	12.99 ± 1.61	0.5^d
Infertility (n, %)	199 (35.2)	145 (35.4)	0.97^e
Infertility duration (months)^a	41.20 ± 29.16	43.23 ± 33.31	0.56^a
Dysmenorrhea (n, %)
No	222 (39.4)	38 (9.3)
Primary	203 (36.1)	180 (44.2)	<0.0001^e
Secondary	138 (24.5)	189 (46.4)	<0.0001^e
Pain scores^a,b,c
Dysmenorrhea	3.92 ± 3.32	6.94 ± 2.72	<0.0001^d
Deep dyspareunia	1.94 ± 3.55	4.46 ± 4.12	<0.0001^d
Non-cyclic chronic pelvic pain	1.88 ± 2.91	3.2 ± 3.23	<0.0001^d
Gastrointestinal symptoms	0.56 ± 1.71	3.81 ± 3.71	<0.0001^d
Lower urinary tract symptoms	0.10 ± 0.8	1.11 ± 2.51	<0.0001^d

^aData are presented as mean ± standard deviation.
^bSometimes more than one for the same patients.
^cVisual analogue scale.
^dStudent t test.
^eχ² test.

Table II. Endometriosis risk according to OC use distribution and the most severe endometriotic lesion type.
	Group A control n = 566	Group B endometriosis type			Adjusted OR (95% CI)^a
	Group A control n = 566	SUP n = 47	OMA n = 120	DIE n = 243	Superficial	Endometrioma	Deep infiltrating
OC user
Never user	160 (28.3)^b	7 (14.9)	27 (22.5)	12 (4.9)
Ever OC user	406 (71.7)	40 (85.1)	93 (77.5)	231 (95.1)	2.59 (1.11–6.03)	1.37 (0.84–2.23)	4.2 (1.54– 11.2)
Current user	142 (25.1)	13 (27.7)	25 (20.8)	41 (16.9)	2.7 (0.98–7.47)	0.95 (0.5–1.7)	1.98 (0.65–6.07)
Past user	264 (46.6)	27 (57.4)	68 (56.7)	190 (78.2)	2.56 (1.07–6.09)	1.65 (0.99–2.75)	5.7 (2.1–15.7)

OC, oral contraception; SUP, superficial endometriosis; OMA, ovarian endometrioma; DIE, deeply infiltrating endometriosis.
^aOC use was adjusted for age, gravidity, infertility, dysmenorrhea and OC use for primary dysmenorrhea.
^bReference category for odds ratio adjusted on logistic regression.

Table III. Endometriosis risk according to OC use and rAFS stages.
	Group A control	Group B rAFS stage				Adjusted OR (95% CI)^a
	Group A control	I	II	III	IV	I	II	III	IV
(n, %)	n = 566	n = 55	n = 85	n = 112	n = 158
Never OC user	160 (28.3)^b	4 (7.3)	5 (5.9)	14 (12.5)	23 (14.6)
Ever OC user	406 (71.7)	51 (92.7)	80 (94.1)	98 (87.5)	135 (85.4)	4.9 (1.75–13.86)	7.8 (2.84–21.87)	2.73 (1.5–4.9)	2.29 (1.4–3.7)
Current user^a	142 (25.1)	17 (30.9)	23 (27.1)	24 (21.4)	16 (9.6)	4.8 (1.6–14.6)	6.5 (2.2–19.2)	1.8 (0.9–3.7)	0.7 (0.4–1.5)
Past user^a	264 (46.6)	34 (61.8)	57 (67.0)	74 (66.1)	119 (75.8)	5.0 (1.7–14.4)	8.6 (3.07–24.2)	3.2 (1.7–5.9)	3.1 (1.9–5.1)

OC: oral contraception.
^aOC use adjusted for age, gravidity, infertility and OC use for primary dysmenorrhea.
^bReference category for odds ratio adjusted on logistic regression.

Table IV. Endometriosis risk according to the indication for oral contraceptive use prescription.
(A)	Control N = 566	Endometriosis, N = 410			OR adjusted^a
No previous OC use	160 (28.3)^b	46 (11.2)
Previous OC use to treat severe primary DM	37 (6.5)	78 (19)			5.6 (3.2–9.8)
Previous OC use indicated for other indications	369 (65.2)	286 (69.8)			2.6 (1.8–4.1)
(B)	Control N = 566	Endometriosis			OR adjusted^a
		SUP N = 47	OMA N = 120	DIE N = 243	SUP	OMA	DIE
No previous OC use	160 (28.3)^b	7 (14.9)	27 (22.5)	12 (4.9)
Previous OC use to treat severe primary DM	37 (6.5)	6 (12.8)	15 (12.5)	57 (23.5)	3.5 (0.9–13.5)	1.9 (0.8–4.3)	16.2 (7.8–35.3)
Previous OC use indicated for other indications	369 (65.2)	34 (72.3)	78 (65)	174 (71.6)	2.8 (1.1–7.1)	1.3 (0.8–2.1)	6.4 (3.2–12.7)

OC: oral contraception; DM: dysmenorrhea.
^aVariables evaluated on regression: age, gravidity, infertility.
^bReference category for adjusted odds ratio on logistic regression.

Comments

Commenting is limited to medical professionals. To comment please Log-in.

Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.