O&G Forum

OBSTETRICS & GYNAECOLOGY FORUM 2021 | ISSUE 3 | 7 O&G Forum 2021; 31: 06 - 10 ORIGINAL RESEARCH embedment, infection and failure. Expulsion rates are lower during intra-operative placement (3.9-10.9%) as compared to early insertion a er normal delivery (25-30%), but it is slightly higher than the expulsion rate a er interval insertion in most studies. 2,3,6,8,9,12,14-16 Expulsion usually occurs within the rst 3 weeks post-insertion. 3,15 Post-insertion infection is uncommon, ranging between 0 and 0.8%. 9,10,14 e incidence of a vaginal discharge is 7%. 10,16 Failure rates are consistently less than 1%. 6,7,9,16-18 e commonest postinsertion complication a er C/S was febrile morbidity (2%) in an observational study by Singal et al, but most of these women had a hospital stay of less than 4 days. 6 In a study by Chawla et al, malpositioning a er vaginal delivery insertion was 62% versus 28% a er operative insertion. Where malpositioning occurred, adverse e ects were increased and continuation rates decreased. 15 Gross malpositioning may be detected by a strings check, although this is not always the case. Women may return with side-e ects such as menstrual irregularities, pain or unintended pregnancy, or it may be detected by ultrasound. 5 In some situations, malpositioned IUDs may be asymptomatic. Identifying a malpositioned IUD presents an opportunity to o er early repositioning or close follow- up, and thereby lower complication rates and improve compliance. 13;16 Although no consensus exists regarding the threshold measurement that would classify an IUD as being correctly placed, most studies describe a distance less than 15 mm from the fundus as acceptable. It must be positioned linearly in the uterine cavity with the horizontal arms reaching laterally towards the cornua. 9,15 Routine scanning at follow-up is not the norm, but all women who are symptomatic or where the strings cannot be seen, should have an ultrasound to locate the IUD. 13;16;19 e incidence of lost strings a er intra-operative IUD insertion can be as high as 50 - 70%. 7,9,14,20 is can be explained by the method of insertion – the strings usually are not introduced into the cervical canal at the time of C/S. It is also dependant on the type of IUD used – e.g. the Cu380A has short strings, whereas the Nova T has long strings. e majority (91.8%) of lost strings can easily be found in the cervical canal. 17 Guiding the strings into the cervical canal and using devices with longer strings can reduce the incidence of lost strings. Some authors recommend the immediate availability of ultrasound at follow- up when intra-caesarean IUD placement is o ered. 14 Even though y-four percent of women in South Africa use some form of contraception, 62% of their last pregnancies was unwanted or unplanned. 21,22 is indicates the still unmet need for the correct method of contraception being used. e reported use of the IUD in South Africa is 1.2%. 21 is is much lower than the use of shorter-acting methods, even though the risk of unintended pregnancy is twenty times higher with short-acting methods. 23 Barriers to IUD insertion included concerns regarding pelvic in ammatory disease, misinformation, fear of the insertion procedure and lack of su cient knowledge and training amongst health care sta . 24 Postpartum IUD insertion is a relatively new practice in the Western Cape, with insertion at C/S increasing around 2014, and a er vaginal delivery around 2015. With rising C/S rates locally and worldwide, intra-caesarean IUD insertion provides the ideal opportunity to enable women to achieve healthy inter-pregnancy intervals. ere are currently many IUDs inserted during C/S, but little is known about the long-term outcomes in the Metro West area. Up till now, surprisingly little data has been published on IUD continuation rates and the factors in uencing it. Follow-up and self-audit are vital in measuring the success of an IUD program and improving outcomes. Challenges at follow-up include poor attendance rates, varying healthcare provider skills, time and nancial constraints in conducting telephonic interviews, ever- changing patient contact details, and the management of missing strings including availability of thread-retrievers and imaging. Where strings are missing, a second visit may be needed as ultrasound machines and/ or operator experience may be lacking. 2;3;5;7;8;10;14;16 In the study by Levi et al, only 48% of women returned for follow-up. Of these women, 72% required an ultrasound, and 26% did not return for their ultrasound appointment. Only 47% of patients were reachable by phone, email or in person by 6 months. 2 e aim of this study is to assess the practice of placement of IUDs at C/S and describe follow-up, with a view to compile best practice guidelines for insertion and follow-up in our clinic setting in the Metro West area of Cape Town. Methods A retrospective descriptive audit of clinical records was performed of all women who received an IUD at C/S between January and June 2018 at Mowbray Maternity Hospital (MMH) and New Somerset Hospital (NSH) in Cape Town. Women with chorioamnionitis, puerperal sepsis, premature or prolonged rupture of membranes and postpartum haemorrhage (PPH) at the time of insertion were excluded. e available IUD in the state sector at the time was the Nova T 380 (Bayer). eatre registers were used to obtain the names and le numbers of all patients who received the IUD during emergency and elective C/S. Obstetric records were obtained and patient demographics and clinical details recorded (tables I and II). e folder numbers and names were cross checked at Groote Schuur Hospital (GSH) and NSH to assess whether the women attended any follow-up visits at the relevant hospitals (Patients who had the IUD inserted at MMH followed up at GSH, and patients who received the IUD at NSH followed up at NSH). Follow-up ndings were recorded. e aim was a sample size of 63 patients attending follow-up at each hospital. is sample size was based on a follow-up rate of 20%, a 95% con dence interval and a power of 80% if 60 IUDs were placed per month at each hospital. Data was entered on a Microso Excel spread sheet a er removal of names and folder numbers and imported into a statistics so ware package (Stata version 13.1, Copyright 1985-2013 Statacorp, LP, USA) for analysis. Institutional consent was obtained from all relevant hospitals, as well as the Western Cape Department of Health (Ref no WC_201810_013). Ethics approval was obtained from the Health Science Faculty Human Research Ethics Committee of the University of Cape Town (Ref no 542/2018). Table I. Patient demographics Demographics n (%) Demographics n (%) Age n = 467 Income 15-19 33 (7.1) Unknown 2 (0.43) 20-24 120 (25.7) < R70 000 p/a 415 (88.9) 25-29 165 (35,3) R 70 0000 - R250 000 p/a 33 (7.1) 30-34 102 (21.8) > R250 000 p/a 13 (2.8) 35-39 43 (9.2) Private 4 (0.9) 40+ 4 (0.9) Marital status Employment Unknown 4 (0.9) Unknown 17 (3.6) Single 284 (60.8) Employed 155 (33.2) Married 137 (29.3) Unemployed 295 (63.2) Partner 42 (9) CLINICAL PATIENT DATA n (%) Parity post-delivery n=467 1 107 (22.9) 4 36 (7.7) 2 176 (37.7) 5 4 (0.9) 3 139 (29.8) 6 1 (0.2) Unknown 4 (0.9) Caesarean section Elective 184 (39.4) Emergency 276 (59.1) Unknown 7 (1.5) Booking Unbooked 2 (0.4) Booked 458 (98.1) Unknown 7 (1.5) CLINICAL PATIENT DATA n (%) Previous CS 1 187 (40.0) 2 73 (15.6) Morethan2 4 (0.9) None 195 (41.8) Unknown 7 (1.5) HIV status Unknown 2 (0.4) Positive 130 (27.8) Negative 335 (71.7) Viral load Unknown 9 (6.9) <1000 111 (85.4) >1000 10 (7.7) Table II: Clinical patient data