Introduction

In the United States, 40% to 57% of women report sexual activity by the 6-week postpartum visit, putting them at risk for unintended pregnancy and short interpregnancy intervals. Unintended pregnancies are associated with increased risk of maternal depression and anxiety, and short interpregnancy intervals are associated with an increased risk of maternal and infant morbidity and mortality, including preterm birth, abruption, preterm premature rupture of membranes, and low birthweight. ^, To optimize interpregnancy intervals and reduce unintended pregnancies, effective contraception should be addressed and provided in the postpartum period.

Long-acting reversible contraception (LARC), which includes intrauterine devices (IUDs) and the etonogestrel subdermal contraceptive implant, are the most effective forms of reversible contraception. LARC use has steadily increased in the United States from about 2% in 2002 to 14% in 2014. LARC use is especially prevalent in the postpartum period, with use reported as 15% to 25% in the first 2–6 months postpartum. ^, Moreover, women who choose IUDs and implants tend to have the highest contraceptive continuation rates.

Timing of postpartum IUD placement varies among providers. Placement may be delayed because of concerns for uterine perforation at the postpartum visit. This results in additional clinic visits, which may reduce IUD uptake while increasing unintended pregnancies and overall costs. One large European prospective cohort study reported a 6-fold increased risk in perforation (6/1000) associated with breastfeeding for postpartum IUD insertions. However, a 2-fold risk persisted up to 36 weeks postpartum compared with a baseline risk of 1 in 1000. Overall, uterine perforation risk was low at <1% despite breastfeeding status. It is neither practical nor cost effective to delay IUD insertion to 36 weeks when risk returns to baseline. Moreover, interval IUD insertion at 4 weeks postpartum and beyond is category 1 (no restrictions) according to the Centers for Disease Control and Prevention (CDC) US Medical Eligibility Criteria for Contraceptive Use (MEC).

Studies comparing the risk of postpartum IUD perforation were primarily performed in the 1980s and predominantly evaluated devices that are no longer available in the United States. ^, The primary objective of this study is to compare uterine perforation rates with postpartum IUD insertion at 4–8 weeks postpartum vs 9–36 weeks postpartum. The secondary objective is to compare uterine expulsion rates for IUDs inserted at these postpartum intervals. We chose these specific intervals to capture the 6-week postpartum visit in the first interval and to include the earliest time that IUDs are considered category 1 by the CDC MEC. We hypothesized that the rate of uterine perforation at the time of postpartum IUD insertion would be greater in the 4–8 weeks postpartum period than the 9–36 weeks postpartum period but that the difference would not be clinically significant.

Materials and Methods

We performed a retrospective cohort study with the Kaiser Permanente Southern California electronic medical record (EMR) from September 2010 to December 2016. Data collection started in 2010 when the specific diagnosis codes used to detect complications existed in the EMR. Data were extracted up to 2016 to allow for at least 1-year follow-up. This study was approved by the Kaiser Permanente Institutional Review Board. Inclusion criteria were the following: women who were at least 18 years old, had an IUD inserted between 4 and 36 weeks after delivery of an infant 24 weeks gestational age or greater and had follow-up for a year after placement. Patients were excluded if they delivered multiples or experienced uterine rupture. Patients identified by billing data as having a complication were manually reviewed to confirm the outcome diagnosis of uterine perforation or IUD expulsion. Because there were no specific International Classification of Diseases, Ninth or Tenth Revision (ICD 9 or 10) codes for uterine perforation caused by the IUD, we identified participants using ICD 9 and 10 codes for mechanical complication owing to intrauterine contraceptive device (996.32, T83.39), genitourinary complications owing to other implant and internal device (996.76), mechanical complication owing to other implant and internal device, (996.59) mechanical complication of genitourinary device (996.3), foreign body in uterus (939.1, T19.3), and displacement of intrauterine contraceptive device (T83.32). All data from manually reviewed patients were assessed for accuracy with double data entry.

Our primary outcome of uterine perforation was defined as laparoscopic- or imaging-confirmed perforation with any portion of the IUD noted beyond the endometrium or a provider-suspected perforation after sounding to a greater than expected depth. We subcategorized perforation by type as uterine (partial), uterine (complete), cervical, and provider-suspected perforation with sounding. Our secondary outcome of expulsion was subcategorized as partial or complete. Partial expulsions were defined as any part of the IUD noted within the cervix either visually on exam or on imaging.

In our manual chart review, we collected information on any pregnancy diagnosis, timing and setting of the IUD complication diagnosis, presenting symptom, and procedure to remove the IUD. We validated data that were ascertained from the EMR regarding weeks postpartum at the time of IUD placement, breastfeeding status, parity status (primiparous or multiparous, which was defined as 2 or more deliveries), type of most recent delivery, type of IUD, and provider type (attending, midlevel, or resident). General demographic data were collected for all patients, including age, body mass index (BMI), race/ethnicity, parity, number of vaginal deliveries, number of cesarean deliveries, most recent type of delivery, breastfeeding status, provider type, and type of IUD placed.

We calculated and compared the proportion of (1) perforations and (2) expulsions with postpartum IUD insertion at 4–8 weeks vs 9–36 weeks, diagnosed within 1 year of placement. The denominator to calculate this rate was the number of postpartum IUD insertions within the specified time frames. Secondarily, we analyzed the data using a multivariate logistic regression model to compare the odds of having a perforation and expulsion between the 2 groups to control for potential confounders. These potential confounders were race/ethnicity, breastfeeding status, IUD type, provider type, most recent delivery type, and BMI.

We calculated our minimum sample size a priori, with the following assumptions. Given the reduced barrier for IUD placement if placed at the postpartum visit, we estimated that patients may still opt for IUD insertion at 4–8 weeks postpartum if the complication rate was low at 1% or less. Although IUD perforation is a rare event occurring at an overall incidence of 1.1–1.4 per 1000, it has been estimated to be higher in the postpartum breastfeeding population. In a recent prospective cohort study, the incidence was 5.6 per 1000 at ≤36 weeks postpartum (this decreases to 1.6 per 1000 at >36 weeks postpartum). For our minimum sample size calculation, to detect a difference of 0.5% in the IUD perforation rate, with a baseline complication rate of 0.5% for the 9 to 36 week postpartum IUD placement group, 80% power, and 5% alpha, we estimated we needed at least 4221 women per group (4–8 weeks and 9–36 weeks), for a total of 8442 women in total. The intervals of 4–8 and 9–36 weeks were chosen to capture the 6-week postpartum visit in the first interval, with a 2-week margin before and after 6 weeks.

For our bivariate analyses, P values for comparing proportions were computed using chi-squared test or Fisher’s exact test, as appropriate based on cell size. Differences were considered statistically significant if P <.05. Any continuous outcomes were compared using t tests or Wilcoxon rank-sum tests, as appropriate based on normality of the data. We performed an unadjusted and adjusted multivariable regression model to determine the odds of a uterine perforation with IUD placement at 4–8 weeks compared with 9–36 weeks postpartum. The primary predictor was placement of the IUD at 4–8 weeks vs >9–36 weeks. We included covariates chosen a priori that have demonstrated clinical significance in the prior literature and/or are biologically plausible, which includes breastfeeding status, most recent delivery type, BMI (≤30 or ≥30), provider type, and IUD type. We added race/ethnicity to our regression model posthoc owing to apparent differences between the 2 groups. An additional analysis was performed to look at the probability of IUD perforation through 36 weeks postpartum utilizing a Kaplan-Meier survival curve. All data were analyzed using Stata Statistical software: Release 15 (StataCorp LLC, College Station, TX).

Results

A total of 24,959 patients met the inclusion criteria. We proceeded with this entire sample to optimize the power of our study to determine and compare the proportion of IUD perforations (a rare complication) and to allow for more variables to be included in our regression models. Of these patients, 841 were identified as having a potential complication. After manual review, 430 patients had the confirmed outcome diagnoses of uterine perforation or expulsion (51%).

There were 13,180 patients in the 4 to 8 week group and 11,777 patients in the 9–36 week group. Mean age, BMI, and IUD type were similar between the groups ( Table 1 ). There was a higher proportion of cesarean deliveries in the 9–36 week group (22.9% vs 26.4%; P <.001). Breastfeeding status was higher in the 4–8 week group (59.3% vs 57.7%). There was also a higher proportion of Hispanic women in the 4–8 week group (60.3 % vs 53%; P <.001) and a higher proportion of White women in the 9 to 36 week group (21.2% vs 27.1%; P <.001).

Perforation was significantly higher with placement at 4–8 weeks postpartum than at 9–36 weeks postpartum (0.78% vs 0.46%; P =.001) ( Table 2 ). Unadjusted expulsion rates were not significantly different between the 2 groups (1.02 vs 1.17; P =.52).

Table 2

Proportion of participants with complications by group

Complication	4–8 wk PP (n=13,180)	9–36 wk PP (n=11,777)	P value a
Perforation (any)	103 (0.78)	54 (0.46)	.001
Uterine, partial	31 (0.24)	16 (0.14)	.07
Uterine, complete	66 (0.50)	32 (0.27)	.004
Cervical	0 (0)	3 (0.03)	.07
With sounding	6 (0.05)	3 (0.03)	.41
Expulsion	135 (1.02)	138 (1.17)	.52
Partial	84 (0.64)	84 (0.71)	.46
Complete	51 (0.39)	54 (0.46)	.38

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. A principled approach to mediation analysis in perinatal epidemiology
2. Cost-effectiveness of agents used for evaluation of ureteral patency during intraoperative cystoscopy in gynecologic and urogynecologic surgery
3. Fetal head descent assessed by transabdominal ultrasound: a prospective observational study
4. Exome sequencing vs targeted gene panels for the evaluation of nonimmune hydrops fetalis

Stay updated, free articles. Join our Telegram channel

Join