Objective
The objective of the study was to determine whether vaginal preparation with povidone iodine before cesarean delivery decreased the risk of postoperative maternal morbidities.
Study Design
The design of the study was a randomized, controlled trial in women undergoing cesarean delivery with subjects assigned to have a preoperative vaginal cleansing with povidone iodine or to a standard care group (no vaginal wash). The primary outcome was a composite of postoperative fever, endometritis, sepsis, readmission, wound infection, or complication.
Results
There were 155 vaginal cleansing subjects and 145 control subjects. Overall, 9.0% developed the composite outcome, with fewer women in the cleansing group (6.5%) compared with the control group (11.7%), although the difference was not statistically significant (relative risk, 0.55; 95% confidence interval, 0.26–1.11; P = .11). Length of surgery, being in labor, and having a dilated cervix were all associated with the composite morbidity outcome.
Conclusion
Vaginal cleansing with povidone iodine before cesarean delivery may decrease postoperative morbidities, although the reduction is not statistically significant.
Cesarean section is the most common surgical procedure performed by obstetricians. One of the most frequent complications of cesarean delivery is infectious morbidity. Endometritis complicates the postoperative course of a cesarean delivery 6-27% of the time. Postcesarean endometritis, up to 10 times more frequent than after vaginal delivery, can lead to serious complications such as bacteremia, peritonitis, intraabdominal abscess, and sepsis. Additionally, cesarean deliveries are frequently complicated by wound morbidities including seroma, hematoma, infection, and separation. These morbidities can lead to significant pain and discomfort, delay in a return to normal function, longer hospital stay, and possible readmission for the newly postpartum woman.
Postpartum febrile morbidity is associated with length of ruptured membranes, length of labor, and number of vaginal examinations. The causal organisms in endometritis and infectious morbidity often arise from ascending inoculation of the upper genital tract by cervicovaginal flora. These bacteria have been demonstrated to be responsible for failure of antibiotic prophylaxis during cesarean deliveries. Additionally, some prophylactic antibiotic regimens are not consistently effective, and the vagina may become colonized with resistant organisms following surgical propylaxis.
Previous studies evaluated whether vaginal preparatory cleansing can reduce the incidence of postoperative infectious morbidity. Povidone iodine, chlorhexidine, and vaginal metronidazole use have been reported. Three studies of povidone iodine had varying results, whereas a study using chlorhexidine showed no reduction in endometritis. A study using vaginal metronidazole reported a reduction in endometritis but was limited by a small sample size. Older data comparing iodine-based solutions to chlorhexidine before hysterectomy reported lower morbidity in the iodine group with improved activity against anaerobic pathogens. Currently, there is no clear recommendation for or against vaginal cleansing before cesarean delivery.
The objective of this study was to determine whether cleansing the vagina with a povidone iodine solution before cesarean delivery decreases the risk of maternal infectious morbidities including endometritis and wound complications.
Materials and Methods
The study was a randomized, controlled trial, approved by the local institutional review board. Subjects were enrolled at 3 different hospitals affiliated with the Department of Obstetrics and Gynecology, Indiana University School of Medicine; University Hospital, a tertiary care referral center; Wishard Memorial Hospital, an urban county hospital; and Methodist Hospital, a large urban hospital. Women undergoing cesarean delivery were enrolled beginning September 2006. The study was internally funded.
The study was open to all pregnant women who were to undergo a cesarean delivery over age 18 years and able to give informed consent. Exclusion criteria were allergy to iodine-containing solutions or planned cesarean hysterectomy. Counseling and consent forms were available in both English and Spanish.
Upon making the determination that a patient was to undergo a cesarean delivery, the physician obtained consent for the procedure and the study. Subjects were randomly assigned to allocation group. Randomization was accomplished by computer-generated random number table to allocate the 2 groups based on odd and even number. Replacement randomization was performed if there was more than a 10% difference in the group size for each 100 randomizations.
The assigned group was written on an index card as either “no wash” for the control group or “vaginal wash” for the experimental group. The cards were sequentially numbered and placed into matching sequentially numbered opaque security envelopes and sealed. Upon consent, the next numbered envelope was opened and the patient’s assignment to either the povidone iodine vaginal washing group or no vaginal washing group was revealed.
A sham wash was not performed for the control group because the current standard is for no vaginal cleansing. In addition, a sham water cleansing could potentially wash away microorganisms, diluting the inoculum. For both these reasons, no sham wash was performed. The surgeon and patient were not blinded to the allocation but were not specifically told of the allocation group either. The Consolidated Standards of Reporting Trials guidelines were followed in this protocol. The trial is registered at clinicaltrials.gov # NCT00386477 .
The patient’s nurse was informed about allocation arm and, upon arrival to operating room and after adequate anesthesia, the intervention group received a povidone iodine vaginal cleansing, in addition to the standard abdominal preparation, by trained operating room personnel. A Foley catheter was placed in all cases in sterile fashion. Vaginal cleansing was performed before the abdomen was scrubbed, taking care to change gloves before performing the abdominal cleansing. The vaginal scrub consisted of 3 sponge sticks soaked in 1% povidone iodine in a prepackaged sterile pouch. The vaginal scrub encompassed the vaginal apex to the introitus with attention to the anterior, posterior, and lateral walls including all fornices.
The control group received the standard abdominal scrub. The cesarean delivery was then performed. All women received the standard antibiotic prophylaxis during the procedure, usually a cephalosporin. The cesarean technique and skin closure procedures were left to surgeon preference. Maternal and infant characteristics and demographics were recorded along with the indication for the cesarean, the duration of labor, and surgical characteristics.
At the time of hospital discharge and again at 1 month postpartum, data were collected by a trained research assistant blinded to the allocation group. The Regenstrief Institute’s computerized medical record system and a hand search of postpartum records were performed for demographic characteristics and presence of postoperative fever, endometritis, and early wound complications.
Definitions for outcome were established before reviewing the charts. Fever was defined as temperature of 38°C or greater excluding the first day after cesarean. Endometritis was defined as uterine tenderness plus postoperative fever requiring antibiotic administration. Wound complications were defined as incisional infection requiring antibiotics, separation, seroma, hematoma, or requirement for debridement. These were clinical diagnoses and required documentation by the care provider in the chart. Chorioamnionitis was defined as fever during labor along with clinical signs of infection (fundal tenderness, tachycardia, purulent lochia, etc) that required starting intravenous antibiotics. A composite complication outcome was defined as the presence of any 1 of the following: endometritis; fever; sepsis; readmission; wound infection; or wound complication (seroma, hematoma, separation).
All data were entered into a Microsoft Access database and analyzed using SPSS software (version 16.0; SPSS, Inc, Chicago, IL). An intent-to-treat analysis was performed. The database was deidentified. The association between vaginal preparation and dichotomous outcomes were investigated using the χ 2 test, with continuity corrections where appropriate, to calculate relative risks (RRs) and 95% confidence intervals (CIs). Continuous variables were compared using a Student t test.
Assuming a postoperative endometritis rate of 10%, to reduce this incidence by half (RR, 0.5), we would need 474 women in each group. All calculations were made assuming 80% power and an α = 0.05. Thus, a total recruited sample of 1000 patients was planned, assuming we would need 948 and that we might be unable to obtain follow-up records for up to 5% of patients.
A planned interim safety analysis was performed after 300 subjects were recruited. At that time, it was noted that there was greater difficulty recruiting women to the trial than anticipated. Given the difficulty of recruitment at the time, the trial was stopped early in January 2009 and the efficacy analysis was performed. A post hoc trend analysis demonstrated that, for the difference in the composite morbidity rates found in the results to reach statistical significance, 517 subjects would be needed in each group, essentially the same as our original planned recruitment.
Results
Three hundred women were consented for the trial. The allocation of women is documented in the Figure . Of the 300 who underwent randomization, 145 were allocated to the control group and 155 to the vaginal cleansing group. The characteristics of the 2 groups are given in Table 1 .
| Variable | Control (n = 145) | Vaginal cleansing (n = 155) |
|---|---|---|
| Age, y | 26.8 (5.9) | 26.6 (5.7) |
| Ethnicity, n (%) | ||
| African American | 53 (36.6) | 49 (31.6) |
| White | 34 (23.4) | 40 (25.8) |
| Hispanic | 54 (37.2) | 64 (41.3) |
| Other | 4 (2.8) | 2 (1.2) |
| Medical coverage, n (%) | ||
| County-funded program | 48 (33.1) | 55 (35.5) |
| Medicaid | 86 (59.3) | 80 (51.6) |
| Private insurance | 5 (3.4) | 11 (7.1) |
| Other | 6 (4.2) | 9 (5.8) |
| Hospital of delivery, n (%) | ||
| Methodist Hospital | 19 (13.1) | 18 (11.6) |
| University Hospital | 9 (6.2) | 12 (7.7) |
| Wishard Hospital | 117 (80.7) | 125 (80.6) |
| Parity | 1.6 (1.5) | 1.6 (1.4) |
| Gestational age at delivery, wk | 38.5 (1.6) | 38.2 (2.7) |
| BMI | 33.9 (7.7) | 33.3 (6.0) |
| Diabetic, n (%) | 11 (7.6) | 16 (10.3) |
| Primary cesarean delivery, n (%) | 54 (37.2) | 55 (35.5) |
| Patient was in labor at time of delivery, n (%) | 50 (34.7) | 45 (29.0) |
| Cervix was dilated at time of cesarean, n (%) | 67 (46.2) | 63 (40.6) |
| Membranes ruptured at time of cesarean, n (%) | 42 (29) | 34 (21.9) |
| Length of membranes ruptured, h | 4.9 (18.0) | 3.1 (14.2) |
| Number of pelvic exams before delivery | 2.0 (3.0) | 1.7 (2.9) |
| Any internal monitor present, n (%) | 28 (19.3) | 24 (15.6) |
| Chorioamnionitis, n (%) | 9 (6.2) | 5 (3.2) |
| Received group B streptococcus prophylaxis, n (%) | 20 (13.8) | 21 (13.5) |
| Transverse uterine incision, n (%) | 143 (98.6) | 149 (96.1) |
| Vaginal hand needed for delivery, n (%) | 4 (2.8) | 4 (2.6) |
| Tubal ligation performed at same surgery, n (%) | 41 (28.3) | 47 (30.3) |
| Ring forceps placed through cervix to allow for better lochia drainage, n (%) | 7 (4.8) | 11 (7.1) |
| Suture closure of the skin, n (%) | 76 (53.1) | 83 (53.5) |
| Estimated blood loss, mL | 686 (207) | 684 (230) |
| Preoperative hemoglobin, g/dL | 11.7 (1.2) | 11.7 (1.3) |
| Postoperative change in hemoglobin, g/dL | 1.6 (1.1) | 1.7 (1.1) |
| Surgical time, min | 51.3 (22.1) | 51.4 (18.0) |
| Days in hospital | 2.1 (0.4) | 2.2 (0.8) |
| Birthweight, g | 3261 (533) | 3307 (826) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
