Ideal Timing of Intrauterine Contraception Insertion After Medical Management of First Trimester Incomplete Abortion

Effectiveness of Early Versus Standard Insertion of Post Abortion Intrauterine Contraception After First Trimester Medical Management of Incomplete Abortion in Central Uganda: A Non-Inferiority Randomized Clinical Trial

Background: Unintended pregnancies continue to cause a public health threat in Low and Middle Income countries yet with restrictive abortion laws. Over 40% of these unintended pregnancies end up as unsafe abortions leading to significant maternal morbidity and mortality. With ovulation occurring between 5-10 days after first trimester abortion, 47% of the women conceive shortly afterwards yet nearly 50% of the women never return for follow up.

This study seeks to investigate effectiveness of early insertion of Intrauterine contraception (IUC) (within one week after medical Post abortion care (mPAC)) compared to Standard IUC insertion (between 2-4 weeks post PAC), in Uganda with the ultimate aim of increasing the uptake of IUC post 1st trimester medical management of incomplete abortion.

Hypothesis:

Null hypothesis: Early Insertion of IUC has higher expulsion rates and continued use than standard insertion after medical management among women with first trimester incomplete abortion at six months within a non-inferiority margin of 7.5%.

Alternate hypothesis: Early Insertion of IUC has similar expulsion rates and continued use as standard insertion after medical management among women with first trimester incomplete abortion at six months within a non-inferiority margin of 7.5%.

Methods: A Non-inferiority RCT of 2,076 participants will be conducted in 15 health facilities within Central Uganda. Baseline participant characteristics will be analyzed using descriptive statistics. For continuous variables, statistical tests such as Fisher's exact test, t-test, ANOVA and Wilcoxon-Mann-Whitney test will be used as appropriate. Binary logistic regression model will be used for factors associated with the dependent variable uptake of intrauterine contraception with p-value set <0.05 as level of significance. Intention to treat and per protocol analysis will be used for the RCTs.

Study utility: A greater proportion of women undergoing medical post abortion care for incomplete abortion will receive their intended post abortion IUC within the first days after the PAC treatment compared with routine insertion at a scheduled follow up visit 2-4 weeks later. Though there could be higher expulsion rates in the early versus the standard IUC insertion, continued use at 6 months following abortion will be higher in the early insertion group as compared to the standard insertion group thereby reducing unplanned and unwanted pregnancy.

Rationale for study:

Early IUC insertions post first trimester medical abortion management avails the opportunity of utilizing the limited human resource in low resource setting. Prior studies have shown higher continuation rates and patient satisfaction with early IUC than with the standard insertion between 2-4 weeks after medical PAC. There's also no need for back up contraception with early IUC as implemented for standard IUC.

Objective: To compare the expulsion rates and continuation rates at six months between early IUC and standard IUD insertions among women managed with mPAC after first trimester incomplete abortion.

Null hypothesis: Early Insertion of IUC has higher expulsion and continued use rates than standard insertion after medical management among women with first trimester incomplete abortion at six months within a non-inferiority margin of 7.5%.

Alternate hypothesis: Early Insertion of IUC has similar expulsion and continued use rates as standard insertion after medical management among women with first trimester incomplete abortion at six months within a non-inferiority margin of 7.5%.

Sample size and calculated power: With an assumed 95% "success" (non-expelled IUDs) in both groups with a power of 90 percent and an alpha of 0.05 the investigators would need to randomize 905 women in each group.

Calculating for sample size for cluster randomized sampling, In a meta-analysis by Adams, the median ICC was observed at 0.005. Considering the sample size calculation of 1770, For 15 clusters (units), average cluster size = 1770/15 = 118 Applying the formula, the investigators get DE = 1.585 Overall sample size = 1770 * 1.585 = 1805 Adjusting for a 15% loss to follow up gives a total sample size of 2076

Sample size estimation for the Continuation rates at 6 months:

Shimoni reported six months' utilization at 69% and 60% for the early and standard insertions respectively.

Based on a non-inferiority margin of 7.5%, a sample size of 497 is needed. For 15 clusters (units), average cluster size = 994/15 = 66 Applying the formula, DE = 1.585 Overall sample size = 994 * 1.585 = 1575 Choosing the bigger sample size and 15 percent loss to follow up which is commonly seen in studies on abortion. Thus, investigators will randomize 2,076 women.

Randomization: The randomization sequence will be generated using STATA 12 software package in a ratio of 1:1 allocation using permuted block size of 4, which will be varied randomly. The randomization will be performed by the Bio-statistician who will remain with the randomization list. This process will be accomplished before the study starts. The investigators and research assistants will not access the randomization list.

Allocation concealment: Sequentially numbered, identical, opaque, sealed envelopes prepared by staff otherwise not involved in the trial to conceal the pre-specified intrauterine contraception allocation from clinicians, research personnel, and participants will be used. After screening, women who will be eligible for IUC will be randomized to receive early or standard insertion of IUC by a different team from the one that ensured allocation.

Blinding: The study will be an open label design. Choice of Study Design: The study will be a non-inferiority trial. Enrolment: All women with incomplete first trimester abortions managed medically with Misoprostol will be invited to be included in the study at the initial outpatient consultation. The women will receive detailed oral and written information regarding the study. An informed consent will be signed by the attending physician and the woman before randomization and any other study related activity.

Allocation and treatment: Designated study nurses and doctors will be responsible for recruiting and examining study participants at the outpatient clinic. Eligible women will then be randomized into either: Group 1; early insertion or Group 2; standard insertion.

Women will be free to decide between the Cu-IUD (Nova T, Bayer AG, Berlin, Germany) and the LNG-IUS (Mirena H, Bayer AG, Berlin, Germany).

Medical PAC will be performed according to the evidence based guidelines. Women with first trimester incomplete abortion diagnosed on history taking and clinical examination will be given Misoprostol 400 mcg sublingually as a single dose. Prior to this, two Ibuprofen 200 mg tablets every 8 hours for three days and two Paracetamol 500mg tablets will be taken orally for three days. Additional analgesia will be available whenever needed. As prophylaxis against post abortion endometritis, two Metronidazole 200mg tablets taken orally every 8 hours for five days and one Doxycycline 100mg tablet twice daily for five days will be given to all participants.

Procedure: The woman will be in supine position with her legs in stirrups. A speculum will be inserted into the vagina and a tenaculum placed on the cervix to straighten the uterus. A sound will be inserted when thought appropriate into the uterus to measure the length to the fundus and thereafter the IUC will be placed at the fundus of the uterus. Threads will be cut at 3 centimeters length.

Follow up: All participants will be followed up for at least one year. Follow up visits will be scheduled at two weeks, three, six and twelve months of IUC insertion. At each scheduled visit, the participants will undertake standardized history and physical examination. If a participant misses a scheduled visit by two weeks she will be contacted by phone, reminded of the missed visit and rescheduling of her appointment done. Home visits will be made for participants who will be lost to follow up. At every visit, the women will be assessed clinically for possible development of PID. They will be asked about development of complications and whether they received any medication from other sources. Information about IUD removal, IUD expulsion, heavy or prolonged menstrual bleeding or missed menstrual periods and hormonal side effects related to progesterone will be ascertained. Women will be requested to keep notes on the menstrual changes on all follow up visits.

Analysis plan: Independent variables between the Early vs. Standard insertion groups will be assessed to determine whether they are similar. The database will contain information on demographic variables that will be analyzed by Mann Whitney U-test. Expulsion as a function of time will be analyzed using Cox regression analysis and visualized by Kaplan Meier curves. In addition multivariate regressions that explore influence of parameters on outcome measures and interactions between various covariates and primary and secondary outcomes will be performed using logistic regression analysis for dichotomous parameters and after dichotomization of continuous parameters. An intention to treat analysis as well as per protocol analysis will be performed.

Participant Retention in the study: The research team will ensure that the participants are comprehensively informed of the follow up schedule. A phone call and text message will always be used to remind the participants at least 72 hours prior to their scheduled visits. Transport reimbursements will be given to all study participants at every follow up visit. A readily available contact of a study nurse will be offered to all participants over 24 hours every week to address any issues that they will have with the study intervention or process. A participant will be regarded as loss to follow up if they fail to honor their appointments after one month of every attempt to have them come back fails.

Data security: All databases will be accessed only by the personnel directly involved in the study. The consents and the questionnaires will be kept separately in a lockable research office. Study participants will not be identified by name in the final report.

Data Monitoring: A multidisciplinary committee comprising of a pharmacist, gynecologist, social worker and one member of the IRB will form the Data Safety and Monitoring Board (DSMB). This committee will report to the steering committee which will comprise of the Doctoral committee. To avoid conflict of interest, the DSMB will have no funders or any of the members involved in the implementation of the research. They will have access to the study findings and will report to the steering committee, interim analysis results with recommendations on whether to halt or continue with the study.

Trial Auditing: A committee of experts in the study to oversee the data control process will be constituted. This committee will work with monitors, trial coordinators and investigators. The monitors will also work with the coordinators to ensure that all the required documents like IRB clearance and correspondence forms, case report forms, patient consent forms, study agreements, adverse events forms, study protocols, institution clearance forms and all source documents like patient charts are filed in a binder that can easily be retrieved whenever needed. The monitors will notify the teams at least a fortnight prior to their visit to ensure timely preparations.

Modifications in the Protocol: In cases of any amendments in the study objectives, major administrative changes, sample size, potential benefits or safety concerns, the IRBs will be notified by the study Principal Investigator. Minor amendments like changes in the administration that don't affect the running of the study will also be communicated at the discretion of the steering committee as a memorandum to the IRBs.

Confidentiality: All study related information will be kept secure at the study sites. Patient files and records will be kept in lockable cabins with limited access. Only the steering committee and DSMB will have full access to the information. Patient identifiers will never be kept with the study findings to ensure confidentiality in the study.

Stopping rules: An interim analysis of results will be performed when 50 percent of women have been recruited. If expulsion rates exceed 20 percent, the study will be stopped. As recommended by Korn, when the Hazard ratio equals or exceeds the preselected non-inferiority margin for our study of 7.5% at 50 percent recruitment, DSMB will stop the study.

Adverse Events: Any untoward medical occurrence without regard to the possible causality in a study participant, in our study will be defined as an adverse event. All adverse events will be recorded from the point the participants sign the consent forms to discharge from the hospital.

All serious adverse events will be reported within seven working days to the IRB. Within 24 hours of occurrence of the adverse events, the investigators will set out to establish the relation of the events whether unexplained or unexpected based on the documentation in the consent process or protocol and post abortion intrauterine devices.

Uterine perforation, need for exploratory laparotomy after post abortion intrauterine device insertions, need for blood transfusion following severe hemorrhage, hospitalization over 72 hours, anaphylactic reaction after IUD insertion, life threatening sepsis, or deaths will be regarded as severe adverse events. Participants with adverse events will be managed immediately by the care teams at the different facilities as per the standard operating procedures.

Quality control: The Principal Investigator will carry out a trial run for a month to pre-test all study instruments, streamline the process of enrolment, allocation and follow up at the health facilities prior to starting data collection. All research assistants involved in the study will undergo training on research conduct prior to the study. A coordination team will be set up to supervise the study and offer support. All filled or recorded data will be checked on a daily basis for completeness. A Data safety monitoring board will be formed to monitor patient safety during the implementation of the RCT.

Inclusion Criteria:

• Women eligible for medical management of first trimester incomplete abortion,
• Staying within ten kilometres from the health facility,
• Opting for post abortion IUC with no intention of conceiving within one year,
• Able and willing to comply with the planned follow up.

Exclusion Criteria:

• Women with known allergies to copper or Levonorgestrel IUDs,
• Patients with unsafe or septic abortions,
• Confirmed cervical cancer or with suspicious Pap smear cytology results demanding more work up,
• Hemoglobin level below 9 g/dl,
• Active genital tract infections,
• Coagulopathies,
• Known uterine anomalies,
• Suspected ectopic pregnancies
• Women with mental health issues that make it hard for them to comprehend the study protocols will be offered the standard of care but will be excluded from the study.