Prophylactic antibiotics for manual removal of placenta in vaginal delivery
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
1. To compare the incidence of postpartum endometritis after manual removal of placenta in vaginal delivery in women who received antibiotic prophylaxis and those who did not.
2. To identify the appropriate regimen of antibiotic prophylaxis for this procedure by comparing the incidence of postpartum endometritis after manual removal of placenta in vaginal delivery in women who received different antibiotic regimens (if antibiotic prophylaxis is found to be effective).
Background
Retained placenta is a potentially life‐threatening condition because of the associated risk of haemorrhage, shock and infection as well as complications related to its removal (Chhabra 2002). This condition continues to be responsible for a high number of maternal fatalities worldwide (WHO 1989). Some studies have reported maternal mortality of 5.6% to nearly 10% in rural areas because of retained placenta (Chhabra 2002; Weeks 2001). Mortality due to retained placenta accounted for 3.33% of all mortality from vaginal deliveries (Chhabra 2002). The main clinical consequence of retained placenta is massive, uncontrolled postpartum haemorrhage, requiring immediate intervention (Stones 1993). This consequence may occur in about 10% of cases (Tandberg 1999).
The reported incidence of retained placenta varied from 1% to 5.5% depending on the definition of prolonged third stage of labour, which ranges from 10 to 60 minutes in various reports (Ely 1995; Selinger 1986; Thomas 1983; Weeks 2001). Retained placenta has been reported with an incidence of 3.3% when 30 minutes was used as the cut‐off point (Combs 1991). Combs and Laros identified several variables associated with haemorrhage: estimated blood loss, difference of haematocrit level, rate of dilatation and curettage, and blood transfusion increase when the third stage of labour nears 30 minutes or more (Combs 1991). There is still no definite agreement about the length of time that should elapse in the absence of bleeding before the placenta is removed manually (Cunningham 2001). When the placenta is not separated promptly after delivery of the baby, and if at any time there is brisk bleeding, placenta can be removed by applying pressure to the body of the uterus and lifting the uterus cephalad by the abdominal hand. This manoeuvre is repeated until the placenta reaches the introitus. Pressure on the uterus is then stopped, allowing the placenta to pass through (Prendiville 1988). If this technique is not possible, manual removal is indicated (Cunningham 2001).
Manual removal of the placenta involves inserting one hand through the vagina into the uterus, increasing the likelihood of bacterial contamination in the uterine cavity. Ely 1995 reported that manual removal of placenta significantly increased risk of postpartum endometritis at vaginal delivery. However, Tandberg 1999 and Thomas 1983 found no increased risk of infection following this procedure. Manual removal of the placenta is not generally considered a risk factor for postpartum endometritis (Gibbs 1980; MacLean 1990). Prophylactic antibiotics after manual removal of placenta are routinely recommended by some (Carroli 1991; Loeffler 1995) but not by others (Ely 1995; Tandberg 1999).
There is no general recommendation for the use of antibiotic prophylaxis for this procedure. The World Health Organization recently suggested a single dose of prophylactic antibiotics with ampicillin 2 gm intravenously plus metronidazole 500 mg intravenously or cefazolin 1 gm intravenously plus metronidazole 500 mg intravenously in manual removal of placenta at vaginal delivery (Mathai 2000).
Using the best current available data, this review aims to determine whether prophylactic antibiotics reduce the incidence of endometritis.
Objectives
1. To compare the incidence of postpartum endometritis after manual removal of placenta in vaginal delivery in women who received antibiotic prophylaxis and those who did not.
2. To identify the appropriate regimen of antibiotic prophylaxis for this procedure by comparing the incidence of postpartum endometritis after manual removal of placenta in vaginal delivery in women who received different antibiotic regimens (if antibiotic prophylaxis is found to be effective).
Methods
Criteria for considering studies for this review
Types of studies
All randomized controlled trials comparing antibiotic prophylaxis and placebo or non antibiotic use to prevent endometritis in manual removal of placenta after vaginal delivery.
Types of participants
All pregnant women undergoing manual removal of placenta after vaginal delivery with gestational age more than 22 weeks, or birthweight greater than 500 g (delivery regarded as birth not an abortion according to the International Classification of Diseases, 10th revision (ICD‐10) (WHO 1992).
Types of interventions
Antibiotic prophylaxis after manual removal of the placenta.
Types of outcome measures
We will consider any trials reporting the following outcomes:
Main outcome
Postpartum endometritis (as defined by authors).
Secondary outcomes
-
Puerperal morbidity (defined as temperature 38.0 ºC (100.4 ºF) or higher, the temperature to occur on any 2 of the first 10 days postpartum, exclusive of the first 24 hours, and to be taken orally by a standard technique at least four times daily) (Cunningham 1993)
-
Perineal infection
-
Duration of hospital stay
-
Sepsis
-
Any infection
-
Blood loss
-
Haemorrhage greater than 1000 ml
-
Secondary postpartum haemorrhage
-
Readmission to hospital
Search methods for identification of studies
We will search the Cochrane Pregnancy and Childbirth Group trial register.
The full list of journals and conference proceedings as well as the search strategies for the electronic databases, which are searched by the Group on behalf of its reviewers, are described in detail in the 'Search strategies for the identification of studies section' within the editorial information about the Cochrane Pregnancy and Childbirth Group. Briefly, the Group searches on a regular basis MEDLINE, the Cochrane Central Register of Controlled Trials (CENTRAL) and reviews the Contents tables of a further 38 relevant journals received via ZETOC, an electronic current awareness service.
Relevant trials, which are identified through the Group's search strategy, are entered into the Group's specialised register of controlled trials. Please see Review Group's details for more detailed information.
In addition, we will search the Cochrane Central Register of Controlled Trials (The Cochrane Library) using the following search strategy:
#1 PLACENTA, RETAINED (MeSH)
#2 placenta* AND (retained or retention or remov*)
#3 ANTIBIOTICS (MeSH)
#4 antibiotic*
#5 #1 or #2
#6 #3 or #4
#7 #5 and #6
We will also search MEDLINE (from 1966 to current), EMBASE (from 1980 to current), CINAHL (from 1982 to current) and LILACS (from 1982 to current), adapting the search strategy by selecting appropriate subject headings and/or free text terms.
We will search for additional studies from the references cited in the retrieved articles. We will review abstracts and letters to the editor to identify randomized controlled trials that have not been published. If we identify a randomized controlled trial, we will contact the primary investigator directly to obtain further data. We will review editorials indicating expert opinion, to ensure that no key studies are overlooked for possible inclusion in this review. We will apply no language restrictions.
Data collection and analysis
Three reviewers will undertake the review. C Chongsomchai will conduct the literature search. C Chongsomchai and P Lumbiganon will independently screen the studies found as a result of the search strategy described earlier, discarding the studies that are clearly ineligible but aiming to be overly inclusive rather than risk losing relevant studies. In the process of screening the papers identified, we will not blind to authorship and journal of origin. We will evaluate trials under consideration for methodological quality using the methods described in the Cochrane Reviewers' Handbook (Clarke 2000). We will grade blinding of randomization, intervention and outcome measurement, and completeness of follow up A: adequate; B: uncertain; C: inadequate. Two reviewers will independently assess whether the studies meet the inclusion criteria with disagreement to be resolved by discussion.
M Laopaiboon (a biostatistician) will analyse data using Review Manager 4.2.2 (RevMan 2003). For binary data, we will calculate event rates, relative risk, and their corresponding 95% confidence intervals. For continuous data, we will use mean difference. We will use forest plot, and the chi squared test of heterogeneity to examine heterogeneity of results (using a 10% level of statistical significance and measuring the value of the I2 statistic for degree of inconsistency (Higgins 2003)). If the detected heterogeneity cannot be explained by any clinical or methodological variation, we will use the random effects model to estimate an overall effect of the prophylactic antibiotic after manual removal of the placenta. We will perform sensitivity analysis to evaluate the robustness of the conclusion according to methodological quality of the included trials.
