Keywords
Kleihauer–Betke - fetomaternal bleeding - cesarean delivery - fetomaternal hemorrhage
- neonatal hematocrit
Fetomaternal hemorrhage has been evaluated in numerous obstetrical clinical conditions,
including trauma, amniocentesis, chorionic villus sampling, external cephalic version,
first trimester abortion, and bleeding during pregnancy.[1]
[2]
[3]
[4]
[5]
[6] Fetomaternal bleeding has also been documented following both vaginal and cesarean
deliveries, with some studies suggesting an increased risk with cesarean delivery,
while others have not found this association.[7]
[8]
[9] One possible difference in these studies may lie in whether or not the placenta
is transected during the cesarean delivery process. Through an extensive literature
search using PubMed, Scopus, Medline, and Google Scholar, we found no prior study
that has evaluated this question. If fetomaternal bleeding is greater in transplacental
transection cesarean deliveries, then Rh-negative mothers who deliver Rh-positive
newborns may need supplemental Rh-hyperimmune globulin. In addition, data are also
limited regarding neonatal hematocrit postcesarean delivery in procedures where the
placenta is transected.
We sought to evaluate the rate of fetomaternal bleeding following routine cesarean
delivery compared with cesarean delivery where transplacental transection was needed
to accomplish delivery. The secondary study objective was to evaluate neonatal hematocrits
postdelivery for each group.
Materials and Methods
Our study was a prospective cohort evaluation of cesarean delivery controls (the placenta
was not transected to accomplish delivery) versus cases (transplacental transection
occurred in the delivery process). All pregnant English-speaking patients who entered
labor and delivery and had a cesarean delivery from January 2016 to April 2018 were
eligible. Once consented, a Kleihauer–Betke's (KB) test was collected within 24 hours
of delivery. For laboratory performance, our institution uses Laboratory Corporation
of America (LabCorp) and for quantifying fetomaternal hemorrhage, LabCorp uses the
KB test. Per LabCorp standardization, a test is considered positive with a cutoff
level of 0.05%, suggesting a fetal bleed of 2.5 mL. A neonatal hematocrit was also
obtained on the first day of life. Though various normal ranges for neonatal hematocrits
have been described, values more than 40% do not usually require any specific follow-up.
A value <40% was chosen for our lower level cutoff because at this level further analysis
will occur to make sure the level does not decrease to a point where transfusion is
needed. Data collection included demographics, gestational age at delivery, indication
for cesarean delivery, type of uterine incision, those in labor prior to cesarean,
manual removal of the placenta, and whether the placenta was transected to effect
delivery of the fetus. Patients with any type of hemoglobinopathy were excluded.
A power analysis was performed based on data from the published studies to date that
evaluated the rate of a positive KB test following cesarean delivery. As transplacental
transection during cesarean delivery is not a frequent occurrence, a 2:1 ratio of
routine cesarean versus transplacental transection cesarean was incorporated. For
a power of 80 using an α of 0.05 and an expected rate of a positive KB test to be
three times higher in the transplacental transection cases, a minimum of 90 patients
was required (60 routine cesarean controls and 30 transplacental transection cases).
The study progressed until the 30 transplacental transection cases were obtained.
Cases number 30 and 31 were obtained on the same day in April 2018. The controls were
obtained by November 2017, and no further controls were collected. A large fetomaternal
bleed was considered at 15 mL or greater. As the estimation of the amount of fetal
bleeding is not completely exact, to err on the side of safety, many blood banks use
this cutoff value to determine if more than one vial of Rh-hyperimmunoglobulin should
be administered. Statistics involved the Student's t-test, chi-square test, Mann–Whitney's U-test, Fisher's exact test, and multivariate analysis where appropriate, and a p-value <0.05 was considered significant with all tests considered against a two-sided
alternative hypothesis. This study was reviewed and approved by the Institutional
Review Board of the University of Tennessee Medical Center, Knoxville, TN.
Results
A total of 94 patients were evaluated with 63 routine cesarean controls compared with
31 transplacental transection cases. The demographics and characteristics of the study
population are seen in [Table 1], and no differences were found between the cases and controls except for manual
removal of the placenta. As depicted, the study population was primarily Caucasian
(89%), multiparous (73%), and underwent repeat cesarean delivery (69%).
Table 1
Demographics and characteristics of the 94 cesarean delivery study subjects
|
Category
|
Cesarean with no transplacental transection
|
Transplacental transection cesarean
|
p-Value
|
|
Number
|
63
|
31
|
|
|
Age (y)
|
27.9 ± 4.9
|
28.1 ± 4.8
|
0.85
|
|
Caucasian
|
58 (92%)
|
26 (84%)
|
0.73
|
|
Multiparity
|
47 (75%)
|
22 (71%)
|
0.90
|
|
Delivery gestational age (wk)
|
38.6 ± 1.9
|
38.5 ± 1.8
|
0.81
|
|
Cesarean delivery indication
|
|
Repeat
|
45 (71%)
|
20 (65%)
|
0.66
|
|
Breech
|
5 (8%)
|
2 (6%)
|
0.99
|
|
Abnormality during labor[a]
|
13 (21%)
|
9 (29%)
|
0.52
|
|
Classical uterine incision
|
2 (3%)
|
2 (6%)
|
0.60
|
|
Manual removal of the placenta
|
10 (16%)
|
23 (74%)
|
0.0001
|
a Failure to progress, fetal heart monitor indications, or a combination of the two.
Overall, there were 15 (16%) positive KB tests ([Table 2]) with 9 (14%, 95% confidence interval [CI]: 7–25%) in the routine cesarean controls
compared with 6 (19%, 95% CI: 7–31%) in the 31 transplacental cesarean cases (p = 0.74). There were three large bleeds as defined, two in the controls and one in
the cases (p = 0.99). There was no difference in finding a positive KB test in the multiparous
group (11 of 69, 16%) versus nulliparous group (4 of 25, 16%), p = 0.99.
Table 2
A total of 94 cesarean delivery study subjects with 63 routine cesarean deliveries
compared with 31 transplacental transection cesarean deliveries
|
Category
|
Cesarean with no transplacental transection
|
Transplacental transection cesarean
|
p-Value
|
|
Number
|
63
|
31
|
|
|
Positive Kleihauer–Betke
|
9 (14%, 95% CI: 7–25)
|
6 (19%, 95% CI: 7–31)
|
0.74
|
|
Large fetomaternal bleed (>15 mL)
|
2 (3%, 95% CI: 0.4–11)
|
1 (3%, 95% CI: 0.1–17)
|
0.99
|
|
Mean neonatal hematocrit
|
50.4% (±5.9%)
|
48.3% (±8.7%)
|
0.17
|
|
Median neonatal hematocrit
|
51.0% (IQR: 46.1–54.5)
|
49.5% (IQR: 40.4–56.5)
|
0.28
|
|
Neonatal hematocrit range
|
36.0–61.6%
|
32.9–63.4%
|
|
|
Neonatal hematocrit <40%
|
2 (3%, 95% CI: 0.4–11)
|
7 (23%, 95% CI: 10–41)
|
0.005
|
Abbreviations: CI, confidence interval; IQR, interquartile range.
Though there were more instances of manual removal of the placenta in the cases (74%)
versus controls (16%), this did not affect the rate of a positive KB test. Of the
23 cases, 5 (21.7%) had a positive KB test compared with 2 of 10 controls (20%), p = 0.99. In analyzing all manual removals of the placenta compared with spontaneous
placental delivery, again there was no difference in 7 of 33 (21.2%) versus 8 of 61
(13.1%), p = 0.47. A multivariate analysis was performed using the variables in [Table 1], and none was found to be significant for producing a higher rate of a positive
KB test.
Mean neonatal hematocrits were not different (p = 0.17). Median neonatal hematocrits with interquartile ranges were also not different
with a z-score of 1.08 (p =0.28). However, for neonatal hematocrits <40%, there were two in the controls (3%
CI: 0.4–11%) compared with seven in the transplacental transection group (23%, 95%
CI: 10–41%), which was significant, p = 0.005. However, none of these nine newborns required a blood transfusion postdelivery.
Discussion
This study did not find that a cesarean delivery that requires transplacental transection
to accomplish delivery results in a greater risk for fetomaternal bleeding. However,
a transplacental transection cesarean delivery did result in an increased risk for
a newborn hematocrit to be <40% at delivery.
Based on our literature review, this is the first study to analyze whether a transplacental
transection cesarean delivery increases the risk for a fetomaternal hemorrhage. Feldman
et al[10] reported a positive KB test rate of 18.6% in 199 cesarean deliveries, which is similar
to our rate of 16%. However, their study used a lower KB cutoff value of 1 mL compared
with our cutoff value of 2.5 mL and most of their study population involved cesarean
delivery following labor. These authors also reported a 5% rate of a fetomaternal
bleed of 25 mL or greater which was similar to our rate of 3% (3 of 94 study participants).
Perslev et al[11] reported a fetomaternal hemorrhage rate of 18.4% in 207 elective cesarean deliveries
using flow cytometry, but only 1.4% had a bleed of 5 mL or greater, which was lower
than our rate of 14% (13 of 94 study participants). Lubusky et al[7] also used flow cytometry to assess fetomaternal hemorrhage and though they found
a higher fetomaternal hemorrhage rate in cesarean delivery compared with vaginal delivery,
they only reported a 2.3% rate of a bleed 5 mL or greater with cesarean delivery.
Adeniji et al[8] used the KB test to evaluate the rate of a fetomaternal bleed in 163 deliveries
(102 vaginal deliveries and 61 cesarean deliveries). The rate of a positive KB test
overall was 10.4%, and the positive KB test rate between vaginal (9.8%) and cesarean
delivery (11.5%) was not different. They reported a rate for a bleed of 15 mL or greater
in the cesarean delivery group of 7% (4 of 61), which was higher but not statistically
different from our rate of 3% (3 of 94), p = 0.43. None of these studies examined the relationship of transplacental transection
at the time of cesarean delivery.
Additionally, in our review, we found no study has reported on the newborn hematocrit
following cesarean delivery with transplacental transection. Our study did show a
significantly higher rate of a neonatal hematocrit <40% in these deliveries. A case
of fetal exsanguination with demise was recently reported involving transection of
the placenta at the time of cesarean delivery combined with difficulty in fetal head
delivery further emphasizing a concern in this clinical setting.[12]
Our study did have a higher positive KB test rate in cesarean deliveries that involved
manual removal of the placenta, though the comparison did not reach significance.
This lack of significance, however, could represent a type 2 error, and therefore,
this is an area that needs further study. Leavitt et al[13] demonstrated that draining the fetal blood from the umbilical cord prior to placental
delivery did reduce the rate of a positive KB test (using a lower cutoff value of
0.5 mL of fetal blood) compared with nondrainage, and this could extrapolate somewhat
into the potential effect of manual placental removal. None of our deliveries involved
draining of the umbilical cord or delayed cord clamping.
Likewise, we only had four cesarean deliveries that involved a classical uterine incision,
and therefore, this study cannot answer the effect of this type of uterine entry on
the rate of fetomaternal hemorrhage.
The strengths of our study are the prospective study design that controlled for the
cesarean indication, type of uterine incision, and manual removal of the placenta,
as well as attaining the numbers needed from the power analysis. A study limitation
is the potential for a type 2 error. However, if the respective rates of a positive
KB test rate between routine cesarean delivery and transplacental transection cesarean
delivery were to remain unchanged, more than 10 times the number of cases and controls
would be needed to reach significance. Additionally, as previously discussed, this
analysis cannot answer whether manual placental removal or classical uterine incision
is risk factors.
Finally, our study used the KB test to evaluate for volume of fetomaternal hemorrhage
based on LabCorp protocol. This test is primarily performed because it is inexpensive,
requires no special equipment, and is easy to perform in situations where rapid results
are needed, which is often the case when evaluating for fetomaternal hemorrhage. However,
flow cytometry allows for automated results and may be more accurate overall. A few
studies have reported on comparisons between the KB test and flow cytometry with most
showing flow cytometry to be more accurate; however, the results have not been universal.[14]
[15]
[16]
[17] However, the primary concern with fetomaternal hemorrhage in patients who are Rh-negative
would be bleeds that exceed the threshold of one vial of Rh-hyperimmune globulin.
Both tests appear to be accurate in identifying large fetomaternal bleeds.
Conclusion
A cesarean requiring transplacental transection to accomplish delivery does not appear
to significantly increase the rate of fetomaternal bleeding. However, a transplacental
transection cesarean delivery may result in an increased risk for newborn hematocrits
to be <40% following delivery. Therefore, we recommend that pediatrics be informed
in clinical settings where the placenta was transected at the time of cesarean delivery,
so these neonates can be evaluated.
