- Email: [email protected]
Bakri balloon during cesarean delivery for placenta previa
International Journal of Gynecology and Obstetrics 124 (2014) 118–122
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
Bakri balloon during cesarean delivery for placenta previa Michael M. Beckmann a,b,⁎, Jacqueline Chaplin b a b
Mater Health Services, Brisbane, Australia Mater Medical Research Institute, Brisbane, Australia
a r t i c l e
i n f o
Article history: Received 11 April 2013 Received in revised form 1 August 2013 Accepted 25 October 2013 Keywords: Bakri balloon Cesarean delivery Placenta previa
a b s t r a c t Objective: To determine if the use of a Bakri balloon at cesarean delivery (CD) for placenta previa is associated with a reduced need for additional surgical or pharmacologic measures and less blood loss than usual practices. Methods: In a randomized controlled trial, 52 women undergoing CD for placenta previa were randomly allocated 1:1 into an intervention arm (prophylactic Bakri balloon immediately following placental delivery) or a control arm (use of any usual surgical/pharmacologic measures to achieve hemostasis). The primary outcomes were a clinician’s decision to undertake further intervention to control bleeding, and the difference between preoperative and postoperative hemoglobin levels. Results: Although fewer women in the intervention group required additional measures to achieve hemostasis during CD, the difference between the groups was not significant (relative risk 0.54; 95% confidence interval, 0.19–1.57). The change in hemoglobin level among women in the intervention arm was also similar to that among controls (2.3 g/dL; 95% confidence interval, –4.4 to 8.9). Conclusion: The prophylactic use of a Bakri balloon at CD for placenta previa tended to be of benefit, with no evidence of harm or patient dissatisfaction, but the need for additional medical/surgical measures to control blood loss was not significantly reduced. Australian New Zealand Clinical Trials Registry: ACTRN12613000348752. © 2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Approximately 1 in every 250 births is complicated by placenta previa [1], where the placenta is implanted entirely or partly in the lower uterine segment. Placenta previa may be subclassified using ultrasound scan to be “major” (implanted across the cervix) or “minor” (not implanted across the cervix) [2]. In most instances, a major placenta previa necessitates a cesarean delivery (CD). Given the rising trend toward birth by CD [3,4] and the known association between previous birth by CD and placenta previa [5], this is becoming an increasingly common complication of pregnancy. Although there are risks associated with all births by CD, when performed in the setting of placenta previa, surgery can be very difficult. Given the poor contractility of the lower segment, CD in this situation is associated with increased risks of bleeding both during and after the operation [6]. Additional surgical procedures to control bleeding are sometimes required, including hysterectomy. Approximately 10% of CDs for placenta previa require an intraoperative or postoperative blood transfusion [7]. Strategies that will reduce the chance of excessive intraoperative/postpartum bleeding in this patient group are clearly desirable.
⁎ Corresponding author at: Mater Health Services, Raymond Terrace, South Brisbane, Queensland 4101, Australia. Tel.: +61 7 31631594; fax: +61 7 3163 1949. E-mail address: [email protected] (M.M. Beckmann).
A Bakri Postpartum Balloon (Cook Medical; Bloomington, IN, USA) is an intrauterine balloon positioned in the uterus and inflated with saline. Its effect may be attributable to a pressure (tamponade) effect on the endometrial cavity or alternatively result from increased hydrostatic pressure in close proximity to the uterine arteries [8]. The catheter also allows drainage of blood to enable clinicians to measure blood loss. There is reasonable evidence supporting its use in the management of postpartum hemorrhage, often when other measures have failed and conservative management is warranted [9,10]. Balloon tamponade catheters have been used in the setting of a CD for placenta previa [11]. There are, however, very limited data specifically supporting the use of the Bakri balloon in the setting of a CD for placenta previa (5 case series including a total number of 26 women [12–16]), and there are no data with respect to the prophylactic use of the Bakri balloon or another balloon at the time of CD for placenta previa. The use of an intrauterine balloon catheter prophylactically at the time of CD may provide a significant benefit in reducing blood loss and avoiding the need for other more extensive surgical procedures. 2. Materials and methods A randomized controlled trial was undertaken at Mater Health Services, Brisbane, Australia, between October 1, 2009, and November 30, 2012. The trial received ethics approval from the Mater Health Services Human Research Ethics Committee. Potential participants were identified by the research midwife and, following a consultation
0020-7292/$ – see front matter © 2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijgo.2013.08.010
M.M. Beckmann, J. Chaplin / International Journal of Gynecology and Obstetrics 124 (2014) 118–122
with an obstetrician or obstetric registrar, eligible women were provided written information about the trial. All women with a live singleton pregnancy at or beyond 24 + 0 weeks undergoing CD in the setting of a known low-lying placenta/placenta previa (leading edge of placenta less than 20 mm from the cervical os) were considered eligible for inclusion in the study. An ultrasound scan was required to confirm the placental location a maximum of 2 weeks prior to the planned CD date. Women with a suspected placenta accreta, congenital abnormalities of the uterine cavity, or submucosal uterine fibroids greater than 5 cm and those aged less than 18 years were ineligible to enroll. The participants were randomly allocated 1:1 into an intervention arm (prophylactic use of the Bakri balloon immediately following delivery of the placenta at the time of CD) or a control arm (use of any/all usual surgical and/or pharmacologic measures to achieve hemostasis during CD). Randomization into the 2 study arms was according to a computer-generated random allocation list. Sealed opaque envelopes were prepared by Mater Medical Research Institute, Brisbane, Australia, and opened by the research midwife after obtaining written consent. Given the nature of the proposed intervention (presence or absence of an intrauterine balloon device), it was not possible to blind either clinicians or women to their allocation. Following delivery of the infant at CD, 1 g of cefazolin and 5 IU of synthetic oxytocin were administered intravenously and the placenta was removed by controlled cord traction. Clinicians were instructed to wait until signs of separation before applying traction to the cord, except in the setting of significant bleeding. Following delivery of the placenta, the cavity was manually explored and a synthetic oxytocin infusion (20 U in 1000 mL of Hartmann solution at 250 mL/hour) was commenced and continued until 4 hours after birth. A further dose of cefazolin (1 g) was administered intravenously at 12 hours post-birth. For women randomized to the intervention arm, a Bakri balloon was prophylactically placed immediately following delivery of the placenta at the time of CD. After removal of the stopcock from the inflation port at the end of the device, the balloon was placed (uninflated) in the uterine cavity. With the woman in the Trendelenburg position, and using a long Harrison forceps, the inflation port end of the balloon device was fed through the cervix from above and retrieved by an assistant vaginally. Following the first-layer closure of the uterine incision, 100 mL of sterile water was inflated into the Bakri balloon to ensure the balloon remained in the cavity. After 2-layer uterine closure, the balloon device was further inflated (up to a maximum total instilled volume of no more than 500 mL) until the uterus was considered “firm.” In the event that the clinician was concerned about ongoing bleeding, any and all alternative surgical and/or pharmacologic measures of achieving hemostasis could now be employed. This included (but was not limited to) the use of misoprostol, ergometrine, prostaglandin-F2α, deflating the balloon and suturing the placental bed/B-Lynch/another compression suture, ligation of the uterine/ovarian/internal iliac arteries, and hysterectomy. At conclusion of the CD, the surgeon performed a digital vaginal examination to confirm that the balloon was in the uterine cavity, and a 1-inch vaginal pack soaked with povidone-iodine was placed to reduce the risk of the balloon being expelled into the vagina. The urethral Foley catheter remained in situ until the Bakri balloon was removed. The balloon remained in situ for no less than 17 hours and no more than 23 hours. Subsequent vaginal loss was recorded up to 36 hours post-birth. For women randomized to the control arm, the clinician was free to employ any and all alternative surgical and/or pharmacologic measures of achieving hemostasis after delivery of the placenta (as described earlier), including the option of a Bakri balloon. A 2-layer uterine closure was then performed. The 2 primary outcome measures were a clinician’s decision to undertake further intervention to control bleeding, and the difference between preoperative and postoperative hemoglobin concentrations. Secondary outcome measures were the specific surgical and/or pharmacologic measures of achieving hemostasis, number of units of red blood
119
cells transfused, intraoperative and postpartum estimated blood loss, endometritis within the first 4 weeks postpartum, return to theater during the hospital admission, admission to the intensive care unit, pain scores (in recovery and then 4 times daily until 24 hours after birth), and patient satisfaction. Blood loss was estimated using a combination of measured loss in the Bakri balloon drainage bag and/or intra-abdominal drains, estimated loss as documented by the treating practitioner in the medical record and/or observation sheets, and estimated loss based on descriptions of vaginal loss (scant, light, moderate, heavy) on the woman’s pad, “bluey,” and vaginal pack (as per Bose et al. [17]). The diagnosis of endometritis was made if antibiotics were administered beyond 24 hours post-birth and the treating clinician documented “endometritis” in the case notes; if antibiotics were administered following hospital discharge because of heavy bleeding, discharge, and/or lower abdominal pain; or if antibiotics were administered beyond 24 hours post-birth in the absence of evidence of a urinary tract infection, breast infection, chest infection, or wound infection. Patient satisfaction was determined by survey administered by a different research officer (blinded to treatment allocation) by phone at approximately 7 days post-birth. Assuming a type-1 error of 0.05 and a power of 80%, a sample size of 50 women per group was calculated to detect: (1) a 20% absolute decrease in the proportion of women requiring further hemostatic intervention (25% in the control arm versus 5% in the intervention arm), and (2) a 1-g/dL difference between the preoperative and postoperative hemoglobin concentrations (from 12 g/dL to 10 g/dL in the control arm versus 12 g/dL to 9 g/dL in intervention arm). To recruit this number of women, a 3-year inclusion period was anticipated. After almost 3 years, and having recruited 52 women to the trial, an interim analysis was undertaken and a decision made to cease enrollment, having considered the feasibility of continuing recruitment to reach an adequate sample size. Statistical analyses of primary and secondary end points were performed according to the intention-to-treat principle, using Stata/SE 9.2 (StataCorp; College Station, TX, USA). Proportional data were compared using the χ2 and Fisher exact tests, and continuous data were compared using the t test. P b 0.05 was considered statistically significant. 3. Results Between October 1, 2009, and November 30, 2012, 83 women underwent CD for placenta previa at Mater Mother’s Hospital, and all 83 were identified and considered for enrollment in the present study. Of these, 8 women did not meet the inclusion criteria and a further 21 declined to participate. Following randomization, 2 women were excluded after a subsequent ultrasound scan indicated, and subsequent histology confirmed, a diagnosis of placenta accreta. In total, 52 women undergoing a CD in the setting of placenta previa were included in the present study (Fig. 1). The groups were similar across all baseline variables except parity (Table 1). Significantly more women randomized to the control group were nulliparous than those randomized to the prophylactic Bakri balloon group (48.2% versus 16.0%; P = 0.019). Although there was a trend toward fewer women in the prophylactic Bakri balloon group requiring additional measures to achieve hemostasis during CD, there was no statistically significant difference between the groups (RR 0.54; 95% CI, 0.19–1.57), even when controlled for parity. Post-hoc subgroup analyses were undertaken to better define particular patient cohorts where the data supported the prophylactic use of a Bakri balloon at the time of CD. Although numbers in the subgroups were small, there was a similar trend favoring the prophylactic use of a Bakri balloon in the settings of preterm CD, major placenta previa, and emergency CD. There was no instance of intraoperative blood transfusion and no difference in the change in hemoglobin level between women receiving a prophylactic Bakri balloon compared with controls, overall or in any subgroup (Table 2).
120
M.M. Beckmann, J. Chaplin / International Journal of Gynecology and Obstetrics 124 (2014) 118–122
Enrollment
Assessed for eligibility (n=83)
Excluded (n=31) ♦ Not meeting inclusion criteria (n=8) ♦ Declined to participate (n=23) ♦ Other reasons (n=0)
Randomized (n=52)
Allocation Allocated to intervention arm (n=25) ♦ Received allocated intervention (n=22)
Allocated to control arm (n=27) ♦ Received allocated intervention (n=27)
♦ Did not receive allocated intervention (n=3)
no reason documented (n=1), unexpected placenta accreta (n=2)
Follow-up Lost to follow-up (n=0)
Lost to follow-up (n=0)
Did not complete follow-up survey (n=3)
Did not complete follow-up survey (n=2)
Analysis Analyzed (n=25)
Analyzed (n=27)
♦ Excluded from analysis (n=0)
♦ Excluded from analysis (n=0)
Fig. 1. Flow of participants through the study (CONSORT flow chart [19]).
Similarly, there were no differences between the groups with respect to measured and estimated blood loss, need for transfusion, intensive care unit admission, pain scores, endometritis (up until 4 weeks postpartum), or any of the individual surgical or pharmacologic measures used to achieve hemostasis at the time of CD (Table 3). When surveyed, similar proportions of women in the 2 groups agreed with statements pertaining to their birth experience, intraoperative/ postoperative pain, and recovery (Table 4). Table 1 Baseline characteristics of the study participants.a Characteristic
Prophylactic use of Bakri balloon (n = 25)
Control arm (n = 27)
P value
White Nulliparity Any previous CD Major placenta previa Anterior placenta Unplanned/emergency CD Preterm birth (b37 + 0 weeks) Age, y Body mass indexc Pregnancy duration, wk Preoperative hemoglobin, mg/dL Preoperative hematocrit, %
19 (76.0) 4 (16.0) 7 (28.0) 19 (76.0) 15 (60.0) 9 (36.0) 8 (32.0) 31.8 ± 5.5 24.2 ± 4.8 36.5 ± 3.1 119.2 ± 11.8 35.0 ± 3.4
16 (59.3) 13 (48.1) 8 (29.6) 20 (74.1) 20 (74.1) 9 (33.3) 10 (37.0) 32.6 ± 5.2 24.7 ± 6.5 37.5 ± 2.1 122.6 ± 9.2 35.9 ± 2.8
0.245 0.019b 0.999 0.999 0.378 0.999 0.776 0.612 0.758 0.172 0.245 0.268
Abbreviation: CD, cesarean delivery. a Values are given as mean ± SD or number (percentage). b P b 0.05. c Calculated as weight in kilograms divided by the square of height in meters.
4. Discussion Among women undergoing a CD for placenta previa, the prophylactic placement of a Bakri balloon immediately following delivery of the placenta does not result in a statistically significant reduction in the need for additional medical/surgical measures to control blood loss, nor does it result in a reduction in subjective or objective measurements of blood loss. However, the present study, which recruited only 52 women, is underpowered, and more than one-quarter of the women enrolled had a minor placenta previa only. Overall, there was a trend toward benefit in placing a Bakri balloon prophylactically at the time of CD. Further analysis of the data indicated that there might be particular cohorts of women undergoing CD for placenta previa who benefit to a greater extent. In the subgroup of women with major placenta previa, for example, the use of a prophylactic Bakri balloon was associated with a reduced need for additional measures to achieve hemostasis (RR 0.42; 95% CI, 0.10–1.92). Moreover, in the subgroup of women undergoing a preterm CD for placenta previa, none of the women in the prophylactic Bakri balloon group (compared with half of the women in the control group) required additional measures to control bleeding. There was also no evidence of harm. It could be anticipated that the placement of a foreign body in the uterus for up to 24 hours postpartum may be associated with an increased risk of infection, but all 3 cases of endometritis were in women randomized to the control group. In addition, there was no evidence that having an intrauterine balloon catheter in situ for up to 24 hours post-birth had any negative impact on women’s birth experience or their experience of postoperative pain.
M.M. Beckmann, J. Chaplin / International Journal of Gynecology and Obstetrics 124 (2014) 118–122
121
Table 2 Primary outcome measures among women undergoing CD for placenta previa. Subgroup
All (n = 52) Preterm CD (n = 18) Major placenta previa (n = 39) Unplanned/emergency CD (n = 18)
Need for further surgical and/or pharmacologic measures to achieve hemostasis during CD
Difference between preoperative and postoperative hemoglobin levels, mg/dL
Prophylactic Bakria (n = 25)
Controla (n = 27)
Relative risk (95% CI)
P value
Prophylactic Bakrib (n = 25)
Controlb (n = 27)
Difference (95% CI)
P value
4 (16.0) 0 (0.0) 2 (10.5) 2 (22.2)
8 (29.6) 5 (50.0) 5 (25.0) 4 (44.4)
0.54 (0.19–1.57) N/A 0.42 (0.10–1.92) 0.50 (0.12–2.08)
0.244 0.019c 0.263 0.340
16.3 20.1 16.9 21.4
18.5 20.3 16.7 18.0
2.3 (–4.4 to 8.9) 0.2 (–12.9 to 13.3) –0.2 (–8.2 to 7.8) –3.4 (–14.8 to 7.9)
0.465 0.978 0.962 0.524
± ± ± ±
11.9 12.6 12.6 10.3
± ± ± ±
11.3 13.4 11.3 11.1
Abbreviations: CD, cesarean delivery; CI, confidence interval. a Values are given as number (percentage). b Values are given as mean ± SD. c P b 0.05.
Table 3 Secondary outcome measures (clinical parameters) among women undergoing CD for placenta previa.a Secondary outcome measure
Prophylactic use of Bakri balloon (n = 25)
Control arm (n = 27)
P value
Intraoperative estimated blood loss, mL Postoperative estimated blood loss, mL Number of units of blood transfused Mean VAS pain score 0–12 hours postpartum 12–24 hours postpartum Additional surgical/pharmacologic measures Bakri balloon (if in control arm) Ergometrine Misoprostol Prostaglandin F2α B-Lynch suture Sutures to placental bed Ligation of internal iliac artery Hysterectomy ICU admission Unplanned return to theater Endometritis
918 ± 610 315 ± 222 0.5 ± 1.1
770 ± 513 363 ± 197 0.4 ± 1.1
0.353 0.417 0.965
1.5 ± 1.1 1.4 ± 1.1
1.5 ± 1.3 1.2 ± 1.2
0.921 0.559
N/A 0 (0.0) 1 (4.0) 0 (0.0) 0 (0.0) 2 (8.0) 0 (0.0) 1 (4.0) 2 (8.0) 0 (0.0) 0 (0.0)
8 (29.6) 2 (7.4) 1 (3.7) 1 (3.7) 0 (0.0) 2 (7.4) 0 (0.0) 0 (0.0) 1 (3.7) 0 (0.0) 3 (11.1)
N/A 0.491 0.999 0.999 N/A 0.999 N/A 0.481 0.603 N/A 0.236
Abbreviations: ICU, intensive care unit; VAS, visual analog scale. a Values are given as mean ± SD or number (percentage).
Using a Bakri balloon following delivery of the placenta during a CD for placenta previa makes anatomic and physiologic sense. Indeed, the Bakri balloon (now most commonly used in the management of postpartum hemorrhage) was originally described as a device to manage
hemorrhage secondary to placenta previa accreta during CD [11]. Cesarean delivery for placenta previa is associated with hemorrhage following placental removal because of the poorly contractile nature of the lower uterine segment [18] and reduced muscle wall thickness (in the setting of a previous CD). As such, the uterus is unable to contract adequately and stop the flow of blood from the open vessels. In theory, uterotonic medications would be of limited benefit in this setting. Tamponade balloons are, however, likely to be effective because they achieve hemostasis through direct pressure applied to the myometrium and placental bed, assuming the uterus is able to contract down on the balloon [13]. Having ceased enrollment after recruiting only 52 women, the present study was not sufficiently powered for either of the 2 primary outcome measures. Further clinical trials are needed to be able to demonstrate benefit of this strategy. Considering the relatively long period of recruitment to enroll 52 women, a multicenter trial (or a similarly sized trial and subsequent meta-analysis) would be required. The present study was, however, the largest study of the use of a Bakri balloon in the setting of placenta previa. It is acknowledged that some of the secondary outcomes (including blood transfusion, hysterectomy, and admission to the intensive care unit) are very uncommon, and it is not possible to draw any conclusion regarding the benefit or otherwise of the prophylactic use of a Bakri balloon in these settings. It is also acknowledged that there was no evidence of benefit in any of the objective measures of blood loss (change in hemoglobin concentration, measured blood loss). However, it is possible that this may have been influenced by the differences in blood loss measurement across the 2 groups (for Bakri balloon catheter bag measurements were only available for women receiving the Bakri balloon). It is also plausible that a difference in blood loss may have
Table 4 Patient satisfaction among women undergoing cesarean delivery for placenta previa.a Secondary outcome measure
Prophylactic use of Bakri balloon (n = 25)
Control arm (n = 27)
P value
Felt the atmosphere in the operating room was comfortable (n = 45) The level of pain I experienced during my CD was no less or greater than I expected (n = 45) The level of pain I experienced after my CD was no less or greater than I expected (n = 43) In the operating room, during the surgery, I felt I was able to: Interact with my partner (n = 43) Bond with the child (n = 44) Have a sense of control (n = 45) Communicate with the staff (n = 44) See the child after delivery (n = 46) Hold the child after delivery (n = 46) I am unsatisfied with the way my labor care was provided (n = 48) I knew what the staff were doing during the operation (n = 43) I was not able to nurse my child after delivery (n = 43) During my hospital stay I was able to move around and care adequately for my child (n = 43) I was not able to bond with my child during my hospital stay (n = 43) My delivery went almost exactly as I had hoped that it would (n = 43) I recovered quickly after my CD (n = 43)
21 (95.5) 17 (77.3) 14 (63.6)
22 (95.7) 17 (73.9) 9 (42.9)
0.999 0.999 0.227
19 (90.5) 17 (77.3) 17 (77.3) 21 (95.5) 18 (81.8) 17 (77.3) 1 (4.4) 19 (86.4) 10 (45.5) 14 (63.6) 1 (4.6) 14 (63.6) 20 (90.9)
18 (81.8) 17 (77.3) 15 (65.2) 20 (90.9) 22 (91.7) 28 (75.0) 1 (4.0) 16 (76.2) 5 (23.8) 15 (71.4) 3 (14.3) 15 (71.4) 15 (71.4)
0.664 0.999 0.514 0.999 0.405 0.999 0.999 0.457 0.203 0.747 0.345 0.747 0.132
Abbreviations: CD, cesarean delivery. a Values are given as number (percentage) of women who agreed or strongly agreed with the statements.
122
M.M. Beckmann, J. Chaplin / International Journal of Gynecology and Obstetrics 124 (2014) 118–122
been realized had the Bakri balloon been inflated using an end point of “control of bleeding,” rather than the clinician’s subjective opinion regarding balloon “firmness.” Despite these limitations, having a relatively quick, simple, single preventative strategy that may obviate the need for additional intervention is likely to be of clinical significance to both junior and senior clinicians faced with a sometimes complex and morbid surgical procedure. In conclusion, the prophylactic use of a Bakri balloon following delivery of the placenta at CD for placenta previa showed a trend toward benefit, and there was no evidence of harm or patient dissatisfaction. However, use of the balloon did not result in a statistically significant reduction in the need for additional medical/surgical measures to control blood loss. Acknowledgments Although the study was financed by Cook Medical, who funded the position of a part-time research midwife, Cook Medical had no involvement in study design, data collection, data interpretation, or data analysis. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication. Conflict of interest The authors have no conflicts of interest. References [1] Faiz AS, Ananth CV. Etiology and risk factors for placenta previa: an overview and metaanalysis of observational studies. J Matern Fetal Neonatal Med 2003;13(3):175–90. [2] Royal College of Obstetricians and Gynaecologists. Placenta praevia, placenta praevia accreta and vasa praevia: diagnosis and management. Green-top Guideline No. 27. http://www.rcog.org.uk/files/rcog-corp/GTG27PlacentaPraeviaJanuary2011.pdf. Published January 2011.
[3] Menacker F. Trends in cesarean rates for first births and repeat cesarean rates for low-risk women: United States, 1990-2003. Natl Vital Stat Rep 2005;54(4):1–8. [4] Black C, Kaye JA, Jick H. Cesarean delivery in the United Kingdom: time trends in the general practice research database. Obstet Gynecol 2005;106(1):151–5. [5] Ananth CV, Smulian JC, Vintzileos AM. The association of placenta previa with history of cesarean delivery and abortion: a metaanalysis. Am J Obstet Gynecol 1997;177(5):1071–8. [6] Neilson JP. Interventions for suspected placenta praevia. Cochrane Database Syst Rev 2003;2:CD001998. [7] Wing DA, Paul RH, Millar LK. Management of the symptomatic placenta previa: a randomized, controlled trial of inpatient versus outpatient expectant management. Am J Obstet Gynecol 1996;175(4 Pt 1):806–11. [8] Cho Y, Rizvi C, Uppal T, Condous G. Ultrasonographic visualization of balloon placement for uterine tamponade in massive primary postpartum hemorrhage. Ultrasound Obstet Gynecol 2008;32(5):711–3. [9] Doumouchtsis SK, Papageorghiou AT, Arulkumaran S. Systematic review of conservative management of postpartum hemorrhage: what to do when medical treatment fails. Obstet Gynecol Surv 2007;62(8):540–7. [10] Dabelea V, Schultze PM, McDuffie Jr RS. Intrauterine balloon tamponade in the management of postpartum hemorrhage. Am J Perinatol 2007;24(6):359–64. [11] Bakri YN. Uterine tamponade-drain for hemorrhage secondary to placenta previaaccreta. Int J Gynecol Obstet 1992;37(4):302–3. [12] Ferrazzani S, Guariglia L, Triunfo S, Caforio L, Caruso A. Successful treatment of postcesarean hemorrhage related to placenta praevia using an intrauterine balloon. Two case reports. Fetal Diagn Ther 2006;21(3):277–80. [13] Bakri YN, Amri A, Abdul Jabbar F. Tamponade-balloon for obstetrical bleeding. Int J Gynecol Obstet 2001;74(2):139–42. [14] Albayrak M, Ozdemir I, Koc O, Demiraran Y. Post-partum haemorrhage from the lower uterine segment secondary to placenta praevia/accreta: successful conservative management with Foley balloon tamponade. Aust N Z J Obstet Gynaecol 2011;51(4):377–80. [15] Marcovici I, Scoccia B. Postpartum hemorrhage and intrauterine balloon tamponade. A report of three cases. J Reprod Med 1999;44(2):122–6. [16] Vrachnis N, Iavazzo C, Salakos N, Papamargaritis E, Boutas I, Creatsas G. Uterine tamponade balloon for the management of massive hemorrhage during cesarean section due to placenta previa/increta. Clin Exp Obstet Gynecol 2012;39(2):255–7. [17] Bose P, Regan F, Paterson-Brown S. Improving the accuracy of estimated blood loss at obstetric haemorrhage using clinical reconstructions. BJOG 2006;113(8):919–24. [18] Sinha P, Kuruba N. Ante-partum haemorrhage: an update. J Obstet Gynaecol 2008;28(4):377–81. [19] Schulz KF, Altman DG, Moher D. CONSORT Group. CONSORT 2010 statement: updated guidelines for reporting parallel group randomized trials. Obstet Gynecol 2010;115(5):1063–70.
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
Bakri balloon during cesarean delivery for placenta previa Michael M. Beckmann a,b,⁎, Jacqueline Chaplin b a b
Mater Health Services, Brisbane, Australia Mater Medical Research Institute, Brisbane, Australia
a r t i c l e
i n f o
Article history: Received 11 April 2013 Received in revised form 1 August 2013 Accepted 25 October 2013 Keywords: Bakri balloon Cesarean delivery Placenta previa
a b s t r a c t Objective: To determine if the use of a Bakri balloon at cesarean delivery (CD) for placenta previa is associated with a reduced need for additional surgical or pharmacologic measures and less blood loss than usual practices. Methods: In a randomized controlled trial, 52 women undergoing CD for placenta previa were randomly allocated 1:1 into an intervention arm (prophylactic Bakri balloon immediately following placental delivery) or a control arm (use of any usual surgical/pharmacologic measures to achieve hemostasis). The primary outcomes were a clinician’s decision to undertake further intervention to control bleeding, and the difference between preoperative and postoperative hemoglobin levels. Results: Although fewer women in the intervention group required additional measures to achieve hemostasis during CD, the difference between the groups was not significant (relative risk 0.54; 95% confidence interval, 0.19–1.57). The change in hemoglobin level among women in the intervention arm was also similar to that among controls (2.3 g/dL; 95% confidence interval, –4.4 to 8.9). Conclusion: The prophylactic use of a Bakri balloon at CD for placenta previa tended to be of benefit, with no evidence of harm or patient dissatisfaction, but the need for additional medical/surgical measures to control blood loss was not significantly reduced. Australian New Zealand Clinical Trials Registry: ACTRN12613000348752. © 2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Approximately 1 in every 250 births is complicated by placenta previa [1], where the placenta is implanted entirely or partly in the lower uterine segment. Placenta previa may be subclassified using ultrasound scan to be “major” (implanted across the cervix) or “minor” (not implanted across the cervix) [2]. In most instances, a major placenta previa necessitates a cesarean delivery (CD). Given the rising trend toward birth by CD [3,4] and the known association between previous birth by CD and placenta previa [5], this is becoming an increasingly common complication of pregnancy. Although there are risks associated with all births by CD, when performed in the setting of placenta previa, surgery can be very difficult. Given the poor contractility of the lower segment, CD in this situation is associated with increased risks of bleeding both during and after the operation [6]. Additional surgical procedures to control bleeding are sometimes required, including hysterectomy. Approximately 10% of CDs for placenta previa require an intraoperative or postoperative blood transfusion [7]. Strategies that will reduce the chance of excessive intraoperative/postpartum bleeding in this patient group are clearly desirable.
⁎ Corresponding author at: Mater Health Services, Raymond Terrace, South Brisbane, Queensland 4101, Australia. Tel.: +61 7 31631594; fax: +61 7 3163 1949. E-mail address: [email protected] (M.M. Beckmann).
A Bakri Postpartum Balloon (Cook Medical; Bloomington, IN, USA) is an intrauterine balloon positioned in the uterus and inflated with saline. Its effect may be attributable to a pressure (tamponade) effect on the endometrial cavity or alternatively result from increased hydrostatic pressure in close proximity to the uterine arteries [8]. The catheter also allows drainage of blood to enable clinicians to measure blood loss. There is reasonable evidence supporting its use in the management of postpartum hemorrhage, often when other measures have failed and conservative management is warranted [9,10]. Balloon tamponade catheters have been used in the setting of a CD for placenta previa [11]. There are, however, very limited data specifically supporting the use of the Bakri balloon in the setting of a CD for placenta previa (5 case series including a total number of 26 women [12–16]), and there are no data with respect to the prophylactic use of the Bakri balloon or another balloon at the time of CD for placenta previa. The use of an intrauterine balloon catheter prophylactically at the time of CD may provide a significant benefit in reducing blood loss and avoiding the need for other more extensive surgical procedures. 2. Materials and methods A randomized controlled trial was undertaken at Mater Health Services, Brisbane, Australia, between October 1, 2009, and November 30, 2012. The trial received ethics approval from the Mater Health Services Human Research Ethics Committee. Potential participants were identified by the research midwife and, following a consultation
0020-7292/$ – see front matter © 2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijgo.2013.08.010
M.M. Beckmann, J. Chaplin / International Journal of Gynecology and Obstetrics 124 (2014) 118–122
with an obstetrician or obstetric registrar, eligible women were provided written information about the trial. All women with a live singleton pregnancy at or beyond 24 + 0 weeks undergoing CD in the setting of a known low-lying placenta/placenta previa (leading edge of placenta less than 20 mm from the cervical os) were considered eligible for inclusion in the study. An ultrasound scan was required to confirm the placental location a maximum of 2 weeks prior to the planned CD date. Women with a suspected placenta accreta, congenital abnormalities of the uterine cavity, or submucosal uterine fibroids greater than 5 cm and those aged less than 18 years were ineligible to enroll. The participants were randomly allocated 1:1 into an intervention arm (prophylactic use of the Bakri balloon immediately following delivery of the placenta at the time of CD) or a control arm (use of any/all usual surgical and/or pharmacologic measures to achieve hemostasis during CD). Randomization into the 2 study arms was according to a computer-generated random allocation list. Sealed opaque envelopes were prepared by Mater Medical Research Institute, Brisbane, Australia, and opened by the research midwife after obtaining written consent. Given the nature of the proposed intervention (presence or absence of an intrauterine balloon device), it was not possible to blind either clinicians or women to their allocation. Following delivery of the infant at CD, 1 g of cefazolin and 5 IU of synthetic oxytocin were administered intravenously and the placenta was removed by controlled cord traction. Clinicians were instructed to wait until signs of separation before applying traction to the cord, except in the setting of significant bleeding. Following delivery of the placenta, the cavity was manually explored and a synthetic oxytocin infusion (20 U in 1000 mL of Hartmann solution at 250 mL/hour) was commenced and continued until 4 hours after birth. A further dose of cefazolin (1 g) was administered intravenously at 12 hours post-birth. For women randomized to the intervention arm, a Bakri balloon was prophylactically placed immediately following delivery of the placenta at the time of CD. After removal of the stopcock from the inflation port at the end of the device, the balloon was placed (uninflated) in the uterine cavity. With the woman in the Trendelenburg position, and using a long Harrison forceps, the inflation port end of the balloon device was fed through the cervix from above and retrieved by an assistant vaginally. Following the first-layer closure of the uterine incision, 100 mL of sterile water was inflated into the Bakri balloon to ensure the balloon remained in the cavity. After 2-layer uterine closure, the balloon device was further inflated (up to a maximum total instilled volume of no more than 500 mL) until the uterus was considered “firm.” In the event that the clinician was concerned about ongoing bleeding, any and all alternative surgical and/or pharmacologic measures of achieving hemostasis could now be employed. This included (but was not limited to) the use of misoprostol, ergometrine, prostaglandin-F2α, deflating the balloon and suturing the placental bed/B-Lynch/another compression suture, ligation of the uterine/ovarian/internal iliac arteries, and hysterectomy. At conclusion of the CD, the surgeon performed a digital vaginal examination to confirm that the balloon was in the uterine cavity, and a 1-inch vaginal pack soaked with povidone-iodine was placed to reduce the risk of the balloon being expelled into the vagina. The urethral Foley catheter remained in situ until the Bakri balloon was removed. The balloon remained in situ for no less than 17 hours and no more than 23 hours. Subsequent vaginal loss was recorded up to 36 hours post-birth. For women randomized to the control arm, the clinician was free to employ any and all alternative surgical and/or pharmacologic measures of achieving hemostasis after delivery of the placenta (as described earlier), including the option of a Bakri balloon. A 2-layer uterine closure was then performed. The 2 primary outcome measures were a clinician’s decision to undertake further intervention to control bleeding, and the difference between preoperative and postoperative hemoglobin concentrations. Secondary outcome measures were the specific surgical and/or pharmacologic measures of achieving hemostasis, number of units of red blood
119
cells transfused, intraoperative and postpartum estimated blood loss, endometritis within the first 4 weeks postpartum, return to theater during the hospital admission, admission to the intensive care unit, pain scores (in recovery and then 4 times daily until 24 hours after birth), and patient satisfaction. Blood loss was estimated using a combination of measured loss in the Bakri balloon drainage bag and/or intra-abdominal drains, estimated loss as documented by the treating practitioner in the medical record and/or observation sheets, and estimated loss based on descriptions of vaginal loss (scant, light, moderate, heavy) on the woman’s pad, “bluey,” and vaginal pack (as per Bose et al. [17]). The diagnosis of endometritis was made if antibiotics were administered beyond 24 hours post-birth and the treating clinician documented “endometritis” in the case notes; if antibiotics were administered following hospital discharge because of heavy bleeding, discharge, and/or lower abdominal pain; or if antibiotics were administered beyond 24 hours post-birth in the absence of evidence of a urinary tract infection, breast infection, chest infection, or wound infection. Patient satisfaction was determined by survey administered by a different research officer (blinded to treatment allocation) by phone at approximately 7 days post-birth. Assuming a type-1 error of 0.05 and a power of 80%, a sample size of 50 women per group was calculated to detect: (1) a 20% absolute decrease in the proportion of women requiring further hemostatic intervention (25% in the control arm versus 5% in the intervention arm), and (2) a 1-g/dL difference between the preoperative and postoperative hemoglobin concentrations (from 12 g/dL to 10 g/dL in the control arm versus 12 g/dL to 9 g/dL in intervention arm). To recruit this number of women, a 3-year inclusion period was anticipated. After almost 3 years, and having recruited 52 women to the trial, an interim analysis was undertaken and a decision made to cease enrollment, having considered the feasibility of continuing recruitment to reach an adequate sample size. Statistical analyses of primary and secondary end points were performed according to the intention-to-treat principle, using Stata/SE 9.2 (StataCorp; College Station, TX, USA). Proportional data were compared using the χ2 and Fisher exact tests, and continuous data were compared using the t test. P b 0.05 was considered statistically significant. 3. Results Between October 1, 2009, and November 30, 2012, 83 women underwent CD for placenta previa at Mater Mother’s Hospital, and all 83 were identified and considered for enrollment in the present study. Of these, 8 women did not meet the inclusion criteria and a further 21 declined to participate. Following randomization, 2 women were excluded after a subsequent ultrasound scan indicated, and subsequent histology confirmed, a diagnosis of placenta accreta. In total, 52 women undergoing a CD in the setting of placenta previa were included in the present study (Fig. 1). The groups were similar across all baseline variables except parity (Table 1). Significantly more women randomized to the control group were nulliparous than those randomized to the prophylactic Bakri balloon group (48.2% versus 16.0%; P = 0.019). Although there was a trend toward fewer women in the prophylactic Bakri balloon group requiring additional measures to achieve hemostasis during CD, there was no statistically significant difference between the groups (RR 0.54; 95% CI, 0.19–1.57), even when controlled for parity. Post-hoc subgroup analyses were undertaken to better define particular patient cohorts where the data supported the prophylactic use of a Bakri balloon at the time of CD. Although numbers in the subgroups were small, there was a similar trend favoring the prophylactic use of a Bakri balloon in the settings of preterm CD, major placenta previa, and emergency CD. There was no instance of intraoperative blood transfusion and no difference in the change in hemoglobin level between women receiving a prophylactic Bakri balloon compared with controls, overall or in any subgroup (Table 2).
120
M.M. Beckmann, J. Chaplin / International Journal of Gynecology and Obstetrics 124 (2014) 118–122
Enrollment
Assessed for eligibility (n=83)
Excluded (n=31) ♦ Not meeting inclusion criteria (n=8) ♦ Declined to participate (n=23) ♦ Other reasons (n=0)
Randomized (n=52)
Allocation Allocated to intervention arm (n=25) ♦ Received allocated intervention (n=22)
Allocated to control arm (n=27) ♦ Received allocated intervention (n=27)
♦ Did not receive allocated intervention (n=3)
no reason documented (n=1), unexpected placenta accreta (n=2)
Follow-up Lost to follow-up (n=0)
Lost to follow-up (n=0)
Did not complete follow-up survey (n=3)
Did not complete follow-up survey (n=2)
Analysis Analyzed (n=25)
Analyzed (n=27)
♦ Excluded from analysis (n=0)
♦ Excluded from analysis (n=0)
Fig. 1. Flow of participants through the study (CONSORT flow chart [19]).
Similarly, there were no differences between the groups with respect to measured and estimated blood loss, need for transfusion, intensive care unit admission, pain scores, endometritis (up until 4 weeks postpartum), or any of the individual surgical or pharmacologic measures used to achieve hemostasis at the time of CD (Table 3). When surveyed, similar proportions of women in the 2 groups agreed with statements pertaining to their birth experience, intraoperative/ postoperative pain, and recovery (Table 4). Table 1 Baseline characteristics of the study participants.a Characteristic
Prophylactic use of Bakri balloon (n = 25)
Control arm (n = 27)
P value
White Nulliparity Any previous CD Major placenta previa Anterior placenta Unplanned/emergency CD Preterm birth (b37 + 0 weeks) Age, y Body mass indexc Pregnancy duration, wk Preoperative hemoglobin, mg/dL Preoperative hematocrit, %
19 (76.0) 4 (16.0) 7 (28.0) 19 (76.0) 15 (60.0) 9 (36.0) 8 (32.0) 31.8 ± 5.5 24.2 ± 4.8 36.5 ± 3.1 119.2 ± 11.8 35.0 ± 3.4
16 (59.3) 13 (48.1) 8 (29.6) 20 (74.1) 20 (74.1) 9 (33.3) 10 (37.0) 32.6 ± 5.2 24.7 ± 6.5 37.5 ± 2.1 122.6 ± 9.2 35.9 ± 2.8
0.245 0.019b 0.999 0.999 0.378 0.999 0.776 0.612 0.758 0.172 0.245 0.268
Abbreviation: CD, cesarean delivery. a Values are given as mean ± SD or number (percentage). b P b 0.05. c Calculated as weight in kilograms divided by the square of height in meters.
4. Discussion Among women undergoing a CD for placenta previa, the prophylactic placement of a Bakri balloon immediately following delivery of the placenta does not result in a statistically significant reduction in the need for additional medical/surgical measures to control blood loss, nor does it result in a reduction in subjective or objective measurements of blood loss. However, the present study, which recruited only 52 women, is underpowered, and more than one-quarter of the women enrolled had a minor placenta previa only. Overall, there was a trend toward benefit in placing a Bakri balloon prophylactically at the time of CD. Further analysis of the data indicated that there might be particular cohorts of women undergoing CD for placenta previa who benefit to a greater extent. In the subgroup of women with major placenta previa, for example, the use of a prophylactic Bakri balloon was associated with a reduced need for additional measures to achieve hemostasis (RR 0.42; 95% CI, 0.10–1.92). Moreover, in the subgroup of women undergoing a preterm CD for placenta previa, none of the women in the prophylactic Bakri balloon group (compared with half of the women in the control group) required additional measures to control bleeding. There was also no evidence of harm. It could be anticipated that the placement of a foreign body in the uterus for up to 24 hours postpartum may be associated with an increased risk of infection, but all 3 cases of endometritis were in women randomized to the control group. In addition, there was no evidence that having an intrauterine balloon catheter in situ for up to 24 hours post-birth had any negative impact on women’s birth experience or their experience of postoperative pain.
M.M. Beckmann, J. Chaplin / International Journal of Gynecology and Obstetrics 124 (2014) 118–122
121
Table 2 Primary outcome measures among women undergoing CD for placenta previa. Subgroup
All (n = 52) Preterm CD (n = 18) Major placenta previa (n = 39) Unplanned/emergency CD (n = 18)
Need for further surgical and/or pharmacologic measures to achieve hemostasis during CD
Difference between preoperative and postoperative hemoglobin levels, mg/dL
Prophylactic Bakria (n = 25)
Controla (n = 27)
Relative risk (95% CI)
P value
Prophylactic Bakrib (n = 25)
Controlb (n = 27)
Difference (95% CI)
P value
4 (16.0) 0 (0.0) 2 (10.5) 2 (22.2)
8 (29.6) 5 (50.0) 5 (25.0) 4 (44.4)
0.54 (0.19–1.57) N/A 0.42 (0.10–1.92) 0.50 (0.12–2.08)
0.244 0.019c 0.263 0.340
16.3 20.1 16.9 21.4
18.5 20.3 16.7 18.0
2.3 (–4.4 to 8.9) 0.2 (–12.9 to 13.3) –0.2 (–8.2 to 7.8) –3.4 (–14.8 to 7.9)
0.465 0.978 0.962 0.524
± ± ± ±
11.9 12.6 12.6 10.3
± ± ± ±
11.3 13.4 11.3 11.1
Abbreviations: CD, cesarean delivery; CI, confidence interval. a Values are given as number (percentage). b Values are given as mean ± SD. c P b 0.05.
Table 3 Secondary outcome measures (clinical parameters) among women undergoing CD for placenta previa.a Secondary outcome measure
Prophylactic use of Bakri balloon (n = 25)
Control arm (n = 27)
P value
Intraoperative estimated blood loss, mL Postoperative estimated blood loss, mL Number of units of blood transfused Mean VAS pain score 0–12 hours postpartum 12–24 hours postpartum Additional surgical/pharmacologic measures Bakri balloon (if in control arm) Ergometrine Misoprostol Prostaglandin F2α B-Lynch suture Sutures to placental bed Ligation of internal iliac artery Hysterectomy ICU admission Unplanned return to theater Endometritis
918 ± 610 315 ± 222 0.5 ± 1.1
770 ± 513 363 ± 197 0.4 ± 1.1
0.353 0.417 0.965
1.5 ± 1.1 1.4 ± 1.1
1.5 ± 1.3 1.2 ± 1.2
0.921 0.559
N/A 0 (0.0) 1 (4.0) 0 (0.0) 0 (0.0) 2 (8.0) 0 (0.0) 1 (4.0) 2 (8.0) 0 (0.0) 0 (0.0)
8 (29.6) 2 (7.4) 1 (3.7) 1 (3.7) 0 (0.0) 2 (7.4) 0 (0.0) 0 (0.0) 1 (3.7) 0 (0.0) 3 (11.1)
N/A 0.491 0.999 0.999 N/A 0.999 N/A 0.481 0.603 N/A 0.236
Abbreviations: ICU, intensive care unit; VAS, visual analog scale. a Values are given as mean ± SD or number (percentage).
Using a Bakri balloon following delivery of the placenta during a CD for placenta previa makes anatomic and physiologic sense. Indeed, the Bakri balloon (now most commonly used in the management of postpartum hemorrhage) was originally described as a device to manage
hemorrhage secondary to placenta previa accreta during CD [11]. Cesarean delivery for placenta previa is associated with hemorrhage following placental removal because of the poorly contractile nature of the lower uterine segment [18] and reduced muscle wall thickness (in the setting of a previous CD). As such, the uterus is unable to contract adequately and stop the flow of blood from the open vessels. In theory, uterotonic medications would be of limited benefit in this setting. Tamponade balloons are, however, likely to be effective because they achieve hemostasis through direct pressure applied to the myometrium and placental bed, assuming the uterus is able to contract down on the balloon [13]. Having ceased enrollment after recruiting only 52 women, the present study was not sufficiently powered for either of the 2 primary outcome measures. Further clinical trials are needed to be able to demonstrate benefit of this strategy. Considering the relatively long period of recruitment to enroll 52 women, a multicenter trial (or a similarly sized trial and subsequent meta-analysis) would be required. The present study was, however, the largest study of the use of a Bakri balloon in the setting of placenta previa. It is acknowledged that some of the secondary outcomes (including blood transfusion, hysterectomy, and admission to the intensive care unit) are very uncommon, and it is not possible to draw any conclusion regarding the benefit or otherwise of the prophylactic use of a Bakri balloon in these settings. It is also acknowledged that there was no evidence of benefit in any of the objective measures of blood loss (change in hemoglobin concentration, measured blood loss). However, it is possible that this may have been influenced by the differences in blood loss measurement across the 2 groups (for Bakri balloon catheter bag measurements were only available for women receiving the Bakri balloon). It is also plausible that a difference in blood loss may have
Table 4 Patient satisfaction among women undergoing cesarean delivery for placenta previa.a Secondary outcome measure
Prophylactic use of Bakri balloon (n = 25)
Control arm (n = 27)
P value
Felt the atmosphere in the operating room was comfortable (n = 45) The level of pain I experienced during my CD was no less or greater than I expected (n = 45) The level of pain I experienced after my CD was no less or greater than I expected (n = 43) In the operating room, during the surgery, I felt I was able to: Interact with my partner (n = 43) Bond with the child (n = 44) Have a sense of control (n = 45) Communicate with the staff (n = 44) See the child after delivery (n = 46) Hold the child after delivery (n = 46) I am unsatisfied with the way my labor care was provided (n = 48) I knew what the staff were doing during the operation (n = 43) I was not able to nurse my child after delivery (n = 43) During my hospital stay I was able to move around and care adequately for my child (n = 43) I was not able to bond with my child during my hospital stay (n = 43) My delivery went almost exactly as I had hoped that it would (n = 43) I recovered quickly after my CD (n = 43)
21 (95.5) 17 (77.3) 14 (63.6)
22 (95.7) 17 (73.9) 9 (42.9)
0.999 0.999 0.227
19 (90.5) 17 (77.3) 17 (77.3) 21 (95.5) 18 (81.8) 17 (77.3) 1 (4.4) 19 (86.4) 10 (45.5) 14 (63.6) 1 (4.6) 14 (63.6) 20 (90.9)
18 (81.8) 17 (77.3) 15 (65.2) 20 (90.9) 22 (91.7) 28 (75.0) 1 (4.0) 16 (76.2) 5 (23.8) 15 (71.4) 3 (14.3) 15 (71.4) 15 (71.4)
0.664 0.999 0.514 0.999 0.405 0.999 0.999 0.457 0.203 0.747 0.345 0.747 0.132
Abbreviations: CD, cesarean delivery. a Values are given as number (percentage) of women who agreed or strongly agreed with the statements.
122
M.M. Beckmann, J. Chaplin / International Journal of Gynecology and Obstetrics 124 (2014) 118–122
been realized had the Bakri balloon been inflated using an end point of “control of bleeding,” rather than the clinician’s subjective opinion regarding balloon “firmness.” Despite these limitations, having a relatively quick, simple, single preventative strategy that may obviate the need for additional intervention is likely to be of clinical significance to both junior and senior clinicians faced with a sometimes complex and morbid surgical procedure. In conclusion, the prophylactic use of a Bakri balloon following delivery of the placenta at CD for placenta previa showed a trend toward benefit, and there was no evidence of harm or patient dissatisfaction. However, use of the balloon did not result in a statistically significant reduction in the need for additional medical/surgical measures to control blood loss. Acknowledgments Although the study was financed by Cook Medical, who funded the position of a part-time research midwife, Cook Medical had no involvement in study design, data collection, data interpretation, or data analysis. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication. Conflict of interest The authors have no conflicts of interest. References [1] Faiz AS, Ananth CV. Etiology and risk factors for placenta previa: an overview and metaanalysis of observational studies. J Matern Fetal Neonatal Med 2003;13(3):175–90. [2] Royal College of Obstetricians and Gynaecologists. Placenta praevia, placenta praevia accreta and vasa praevia: diagnosis and management. Green-top Guideline No. 27. http://www.rcog.org.uk/files/rcog-corp/GTG27PlacentaPraeviaJanuary2011.pdf. Published January 2011.
[3] Menacker F. Trends in cesarean rates for first births and repeat cesarean rates for low-risk women: United States, 1990-2003. Natl Vital Stat Rep 2005;54(4):1–8. [4] Black C, Kaye JA, Jick H. Cesarean delivery in the United Kingdom: time trends in the general practice research database. Obstet Gynecol 2005;106(1):151–5. [5] Ananth CV, Smulian JC, Vintzileos AM. The association of placenta previa with history of cesarean delivery and abortion: a metaanalysis. Am J Obstet Gynecol 1997;177(5):1071–8. [6] Neilson JP. Interventions for suspected placenta praevia. Cochrane Database Syst Rev 2003;2:CD001998. [7] Wing DA, Paul RH, Millar LK. Management of the symptomatic placenta previa: a randomized, controlled trial of inpatient versus outpatient expectant management. Am J Obstet Gynecol 1996;175(4 Pt 1):806–11. [8] Cho Y, Rizvi C, Uppal T, Condous G. Ultrasonographic visualization of balloon placement for uterine tamponade in massive primary postpartum hemorrhage. Ultrasound Obstet Gynecol 2008;32(5):711–3. [9] Doumouchtsis SK, Papageorghiou AT, Arulkumaran S. Systematic review of conservative management of postpartum hemorrhage: what to do when medical treatment fails. Obstet Gynecol Surv 2007;62(8):540–7. [10] Dabelea V, Schultze PM, McDuffie Jr RS. Intrauterine balloon tamponade in the management of postpartum hemorrhage. Am J Perinatol 2007;24(6):359–64. [11] Bakri YN. Uterine tamponade-drain for hemorrhage secondary to placenta previaaccreta. Int J Gynecol Obstet 1992;37(4):302–3. [12] Ferrazzani S, Guariglia L, Triunfo S, Caforio L, Caruso A. Successful treatment of postcesarean hemorrhage related to placenta praevia using an intrauterine balloon. Two case reports. Fetal Diagn Ther 2006;21(3):277–80. [13] Bakri YN, Amri A, Abdul Jabbar F. Tamponade-balloon for obstetrical bleeding. Int J Gynecol Obstet 2001;74(2):139–42. [14] Albayrak M, Ozdemir I, Koc O, Demiraran Y. Post-partum haemorrhage from the lower uterine segment secondary to placenta praevia/accreta: successful conservative management with Foley balloon tamponade. Aust N Z J Obstet Gynaecol 2011;51(4):377–80. [15] Marcovici I, Scoccia B. Postpartum hemorrhage and intrauterine balloon tamponade. A report of three cases. J Reprod Med 1999;44(2):122–6. [16] Vrachnis N, Iavazzo C, Salakos N, Papamargaritis E, Boutas I, Creatsas G. Uterine tamponade balloon for the management of massive hemorrhage during cesarean section due to placenta previa/increta. Clin Exp Obstet Gynecol 2012;39(2):255–7. [17] Bose P, Regan F, Paterson-Brown S. Improving the accuracy of estimated blood loss at obstetric haemorrhage using clinical reconstructions. BJOG 2006;113(8):919–24. [18] Sinha P, Kuruba N. Ante-partum haemorrhage: an update. J Obstet Gynaecol 2008;28(4):377–81. [19] Schulz KF, Altman DG, Moher D. CONSORT Group. CONSORT 2010 statement: updated guidelines for reporting parallel group randomized trials. Obstet Gynecol 2010;115(5):1063–70.