Sunteți pe pagina 1din 10

Risk Factors for Severe Postpartum Hemorrhage After

Cesarean Delivery: Case-Control Studies


Alexander J. Butwick, MBBS, FRCA, MS,* Bharathi Ramachandran, BS,* Priya Hegde, BA,*
Edward T. Riley, MD,* Yasser Y. El-Sayed, MD, and Lorene M. Nelson, MS, PhD

BACKGROUND: Women who undergo intrapartum caesarean delivery (CD) are at increased risk
of postpartum hemorrhage (PPH) compared with those undergoing prelabor CD. To determine
whether the presence and strength of the associations between individual risk factors and
severe PPH vary among women undergoing prelabor CD or intrapartum CD, stratified analyses
are needed according to CD subtype.
METHODS: To identify risk factors for severe PPH within 2 distinct CD populations, prelabor CD
and intrapartum CD, we performed 2 case-control studies. Women in each study cohort delivered
at a tertiary obstetric center in the United States between 2002 and 2012. For each study, cases
were women who had a blood loss 1500 mL or who received an intraoperative or postoperative
transfusion up to 48 hours after delivery. Risk factors for severe PPH among women undergoing
prelabor CD or intrapartum CD were examined in separate logistic regression models.
RESULTS: For prelabor CD, we identified 269 cases and 550 controls. Clinical factors with the
highest adjusted odds for severe PPH during prelabor CD were general anesthesia (adjusted
odds ratio [aOR] = 22.3; 95% confidence interval [CI] = 4.999.9; reference group = spinal
anesthesia), multiple pregnancies (aOR = 8.0; 95% CI = 4.215.0; reference group = singleton
pregnancy), and placenta previa (aOR = 6.3; 95% CI = 3.411.8). For intrapartum CD, we iden-
tified 278 cases and 572 controls. Clinical factors with the highest adjusted odds for severe
PPH during intrapartum CD were general anesthesia (aOR = 5.4; 95% CI = 1.717.1), multiple
pregnancies (aOR = 3.2; 95% CI = 1.76.3), and a predelivery hemoglobin 9.9 g/dL (aOR =
3.0; 95% CI = 1.36.9; reference group = predelivery hemoglobin 11 g/dL).
CONCLUSIONS: Women who undergo prelabor CD and intrapartum CD have several shared risk
factors for severe PPH (general anesthesia and multiple pregnancies). However, the risk factor
profiles for severe PPH differed between these CD cohorts. Recognizing these differences may
be important when planning resources and interventions for high-risk patients undergoing either
prelabor or intrapartum CD.(Anesth Analg 2017;XXX:0000)

S
evere postpartum hemorrhage (PPH) is an important undergoing CD after induction and 130% among women
cause of maternal death and severe maternal mor- undergoing noninduced CD.6
bidity.13 Compared with vaginal delivery, women The International PPH Collaborative group, which com-
undergoing cesarean delivery (CD) incur the highest risk prises an international panel of clinical epidemiologists, has
of PPH and hemorrhage-related morbidity.4,5 Furthermore, called for more studies using clinically rich data to better
evidence suggests that PPH during CD is occurring more understand relevant, and potentially preventable, risk fac-
frequently. In the United States, between 1994 and 2006, tors associated with PPH.7 Once risk factors for PPH are
the rate of atonic PPH increased 160% among women identified, clinical predictive rules could be developed.
Implementation of predictive rules into clinical practice
could optimize patient outcomes8,9 by improving the plan-
From the Departments of *Anesthesiology, Perioperative, and Pain Medicine;
Obstetrics and Gynecology; and Health Research and Policy (Division of ning and timely mobilization of staffing and resources
Epidemiology), Stanford University School of Medicine, Stanford, California. before bleeding onset. For example, when an at-risk preg-
Accepted for publication January 12, 2017. nant patient is identified before delivery, providers would
Funding: This study was supported and funded by the Eunice Kennedy have the opportunity to order and prepare blood products,
Shriver National Institute of Child Health and Development, National
Institute of Health. The study was also supported by the Department of obtain additional venous access, and prepare additional
Anesthesiology, Perioperative, and Pain Medicine; and Department of equipment for PPH management. High-risk pregnant
Obstetrics and Gynecology, Stanford University School of Medicine. A.J.B.
was supported by an award from the Eunice Kennedy Shriver National patients could also receive enhanced postpartum surveil-
Institute of Child Health and Development (K23HD070972). lance for excessive blood loss. Finally, for high-risk patients
The authors declare no conflicts of interest. awaiting prelabor CD, arrangements could be made for
Supplemental digital content is available for this article. Direct URL citations delivery in an obstetric center with the necessary staff and
appear in the printed text and are provided in the HTML and PDF versions of
this article on the journals website (www.anesthesia-analgesia.org). resources for providing effective PPH management.
Reprints will not be available from the authors. Identifying risk factors for PPH during CD has been chal-
Address correspondence to Alexander Butwick, MBBS, FRCA, MS, Depart- lenging because key differences exist in patient, obstetric,
ment of Anesthesiology, Perioperative and Pain Medicine (MC: 5640), Stan- and intrapartum characteristics for women who undergo
ford University School of Medicine, 300 Pasteur Dr, Stanford, CA 94305.
Address e-mail to ajbut@stanford.edu.
prelabor CD versus CD after onset of labor or induction of
Copyright 2017 International Anesthesia Research Society labor (hereafter referred to as intrapartum CD). Based on
DOI: 10.1213/ANE.0000000000001962 data from 2 population-wide studies in Norway, the risk

XXX 2017 Volume XXX Number XXX www.anesthesia-analgesia.org


1
Copyright 2017 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Postpartum Hemorrhage and Cesarean Delivery

of PPH is reported to be higher for women undergoing Inclusion criteria for intrapartum CD were women with
intrapartum CD compared with prelabor CD (3.1% vs 2%, documented evidence of painful regular contractions or
respectively).4,10 Although intrapartum factors, such as cho- induction of labor before CD. We excluded women with
rioamnionitis and oxytocin exposure,1113 may explain why known hematologic or coagulation disorders (apart from
the risk of atonic PPH is higher after labor induction, to the those with HELLP [hemolysis, elevated liver enzymes,
best of our knowledge, only 3 studies have investigated risk low platelet count] syndrome) or who were anticoagulated
factors for PPH for women who underwent elective CD before CD. Based on literature review,4,5,10,1719 the criteria for
and nonelective or emergency CD.1416 Only 1 study differ- severe PPH were an EBL 1500 mL after CD or red blood
entiated women by the presence or absence of labor before cell (RBC) transfusion during or within 48 hours after CD.
CD.16 To determine whether the presence and strength of We selected severe PPH as our primary outcome for several
the associations between individual risk factors and severe reasons. First, average EBL values during CD can approach
PPH vary among women undergoing prelabor CD or intra- EBL thresholds for defining PPH. Second, an EBL of 1500
partum CD, stratified analyses are needed according to CD mL may reflect the point where physiological compensation
subtype. starts to fail resulting in greater obstetric morbidity.20 Third,
The primary objectives of this observational study were EBL may often be underestimated. Therefore, to account for
to perform 2 case-control studies to identify independent subjects in whom EBL may have been underestimated, we
predictors for severe PPH within each of the following CD incorporated RBC transfusion within 48 hours of CD as a
subpopulations: prelabor CD and intrapartum CD. classifier for severe PPH. In each CD cohort, 2 controls were
randomly selected with same year of delivery as each case
METHODS using incidence density sampling. Each control had an EBL
This study was approved by the Stanford University IRB < 1500 mL and no transfusion. EBL and transfusion data
(Protocol ID = 25236). We performed 2 case-control studies were reviewed in the medical records to confirm correct
classification of each case and control.
to identify clinical risk factors for severe PPH after prelabor
Based on literature review5,10,1416,2123 and clinical plau-
CD and intrapartum CD. Our study cohorts were identified
sibility, potential risk factors for severe PPH in each CD
from a population who delivered by CD at Lucile Packard
cohort were considered. For both CD cohorts, we abstracted
Childrens Hospital (LPCH), a tertiary care obstetric center,
data from the medical records for the following candi-
between January 2002 and December 2012.
date variables: maternal age, self-reported race/ethnic-
To identify women who underwent CD between January
ity, weight, parity, gestational age at delivery, number of
2002 and December 2006, we used the Stanford Translational
previous CDs, chronic hypertension, gestational diabetes,
Research Integrated Database Environment (STRIDE)
singleton/multiple pregnancies, gestational hypertension,
through a combined search of International Classification
preeclampsia, HELLP syndrome, hemoglobin (Hb) most
of Diseases (ICD-9) procedure codes (74.x) and Current
proximate to delivery, time of CD (classified as weekday
Procedural Terminology codes (01961, 01963, 01968, 59510,
daytime [between 7:00 am and 4:59 pm], weekday night
59515, 59618, 59620, 59622, 1008991, 1014218, and 1014220). time [between 5:00 pm and 6:59 am], and weekend) and
Based on a standards-based informatics platform, STRIDE mode of anesthesia (spinal, epidural, combined spinal-epi-
is a clinical data warehouse that contains linked demo- dural [CSE], or general anesthesia). We designated the final
graphic, administrative, and clinical data (from electronic mode of anesthesia before surgical incision as the mode of
or scanned medical records) from LPCH and Stanford anesthesia in our analysis (data on final anesthesia mode are
University Medical Center. The clinical data warehouse presented in Supplemental Digital Content, Supplemental
can be searched using search queries input by program- Figures 1 and 2, http://links.lww.com/AA/B691). For
mers based in the Stanford Center for Clinical Informatics, example, if, before surgery, a patient received a spinal as the
Stanford University. We identified women with severe PPH primary anesthetic that failed, and then received a second
through a combined search for ICD-9 codes for postpartum blocka CSEwhich provided adequate surgical anesthe-
hemorrhage (666.x). Because an electronic medical record sia, the designated anesthesia mode was CSE. To account
system was implemented at LPCH in 2007, we searched for temporality of the potential relationship between the
a computerized clinical database (LPCH Information and preincision mode of anesthesia with severe PPH, we did not
Knowledge System) to identify CDs between January 2007 include information for patients who required intraopera-
and December 2012 through a search of Medicare Severity tive conversion from neuraxial to general anesthesia in our
Diagnosis Related Group codes 765 and 766. Estimated analyses. For prelabor CD, we abstracted additional data for
blood loss (EBL) and transfusion data for CDs between 2007 the following characteristics: earlier history of dilation and
and 2012 were available in the clinical database. Between curettage (D&C) or dilation and evacuation (D&E), and pre-
2002 to 2006 and 2007 to 2012, there were a total of 6997 CDs vious myomectomy. For CD after labor, we abstracted addi-
and 8192 CDs, respectively. tional data for the following characteristics: spontaneous
Trained research assistants (P.H. and B.R.) reviewed versus induced labor, labor augmentation with oxytocin,
medical charts to determine whether patients met inclu- chorioamnionitis, and cervical dilation before CD. In light of
sion criteria for prelabor CD or intrapartum CD. Inclusion American College of Obstetricians and Gynecologists guide-
criteria for prelabor CD were scheduled CD without labor; lines to avoid nonmedically indicated deliveries before 39
CD following unsuccessful external cephalic version; and weeks of gestation,24 we presumed that some women with a
CD for obstetric or fetal indications, such as preeclamp- history of earlier CD were admitted with spontaneous labor
sia, non-reassuring fetal trace without labor or induction. and underwent intrapartum CD before 39 weeks. Therefore,

2
www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
Copyright 2017 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
we accounted for number of earlier CDs in our regression prevalence for characteristics among controls, and an of
model for intrapartum CD. Because curettage of the uterus .05, we calculated that the smallest odds greater than 1 that
during a D&E or D&C may significantly damage the endo- could be detected with 80% power is an odds ratio of 1.85.
metrium and uterine cavity to influence placental adher- All analyses were conducted using STATA version 12
ence, we only considered this variable in the prelabor CD (StataCorp, College Station, TX). Data are presented as
cohort. Primary indications for prelabor and intrapartum mean (SD), median (interquartile range), or n (%).
CD were also examined (data presented in Supplemental
Digital Content, Supplemental Table 1 and Table 3, http:// RESULTS
links.lww.com/AA/B691) and were stratified by mode of Prelabor Cesarean Delivery
anesthesia. The most common indications for prelabor CD In this cohort, we identified 269 women with severe PPH
were prior CD (56%) and macrosomia (23%). The most com- and 550 matched controls. Women with severe PPH had a
mon indications for intrapartum CD were non-reassuring significantly higher median EBL compared with controls
fetal trace (76%) and labor arrest (32%). (Frequencies total (1600 [15002000] mL vs 800 [600900] mL; P < .001). Among
more than 100% as some patients had more than 1 indica- severe PPH cases, 223 (82.9%) women had at least 1500
tion for intrapartum CD.) mL EBL, and 148 (55%) women received RBC intraopera-
Among intrapartum CDs, we performed subanalyses to tively or within 48 hours after CD. Among all women who
assess rates of severe PPH among women with cervical dila- received RBC, 91 women received RBC intraoperatively and
tion <10 cm who did and did not experience arrest of the first 90 women received RBC within 48 hours after CD, of whom
stage, and women with a cervical dilation of 10 cm who did 33 (36.7%) had an EBL < 1500 mL and did not receive intra-
and did not experience a prolonged second stage of labor. operative RBC. Twenty-four women (19 cases; 5 controls)
Women were classified with arrest of the first stage of labor experienced primary anesthetic block failure before surgery
and prolonged second stage by using criteria described by (Supplemental Digital Content, Supplemental Figure 1,
the American College of Obstetricians and Gynecologists25 http://links.lww.com/AA/B691).
(see Appendix for full descriptions). These guidelines were Maternal, obstetric, intrapartum, and perioperative
in place during the study period. characteristics of women with and without severe PPH are
presented in Table1. Unadjusted and adjusted odds ratios
STATISTICAL ANALYSIS for clinical factors selected for multivariable analysis are
We performed 2 analyses to assess characteristics and risk presented in Table 2. In the multivariable model, factors
factors for (1) women undergoing prelabor CD and (2) independently associated with severe PPH were general
women undergoing intrapartum CD. Within each CD popu- anesthesia (adjusted odds ratio [aOR] = 22.3; 95% confi-
lation, patient and obstetric characteristics for women with dence interval [CI] = 4.999.9; reference group = spinal anes-
and without severe PPH were compared using analysis of thesia), multiple pregnancies (aOR = 8.0; 95% CI = 4.215.0;
variance, Mann-Whitney U tests for continuous variables reference group = singleton pregnancy), placenta previa
and 2 test and Fisher exact test for categorical variables. (aOR = 6.3; 95% CI = 3.411.8), predelivery Hb 9.9 g/dL
For each analysis, the primary outcome was severe PPH. (aOR = 2.8; 95% CI = 1.07.5 [reference group = predelivery
For each continuous variable associated with PPH, we Hb 11 g/dL]), Hispanic race (aOR = 2.4; 95% CI = 1.05.8
used restricted cubic spline functions to test whether the [reference group = non-Hispanic white]), and a history of
associations met the assumptions of a linear relationship. previous D&C or D&E (aOR = 1.9; 95% CI = 1.23.0). Model
Continuous variables were categorized if they were not calibration (HosmerLemeshow test, P = .3) and model dis-
linearly related to the outcome using cubic spline function crimination (AUROC = 0.86; 95% CI = 0.830.89) were good.
plots (maternal age and gestational age at delivery) or clini- After 500 cycles of bootstrapping, the c-index remained
cally relevant cut points (predelivery Hb). Multivariable unchanged.
analysis was performed using multivariable logistic
regression, which included all variables with P < .10 in Intrapartum Cesarean Delivery
the univariable analyses. We used variance inflation factor In this cohort, we identified 278 women with severe PPH
testing to assess collinearity between candidate variables. and 572 matched controls. The mean EBL values were sig-
Collinearity was determined to be insignificant as variance nificantly higher among cases compared with controls (1685
inflation factor (VIF) scores for the prelabor CD ranged from [665] mL vs 781 [202] mL; P < .001). Among severe PPH cases,
1.03 to 1.27 (mean VIF score = 1.14) and for CD after labor 209 (75.4%) women had at least 1500 mL EBL, and 157 (56.5%)
ranged from 1.01 to 1.6 (mean VIF score = 1.21). To account women received RBC intraoperatively or within 48 hours
for the increased probability of making a type I error when after CD. Among all women who received RBC, 51 women
making multiple comparisons, we adjusted the confidence received RBC intraoperatively and 122 women received RBC
intervals and P values for each comparison in our multivari- within 48 hours after CD, of whom 59 (48.4%) had an EBL <
able models using a Bonferroni correction.26 Goodness of fit 1500 mL and did not receive intraoperative RBC. Maternal,
was evaluated using the Hosmer-Lemeshow statistic. We obstetric, intrapartum, and perioperative characteristics are
calculated the area under the receiver-operating character- presented in Table 3. Because earlier clinical studies have
istic curve (AUROC) using standard methods.27 We recalcu- observed an association between labor augmentation with
lated the c-index for each model using 500 iterations of the PPH,11,23,29,30 we forced labor augmentation with oxytocin into
Efron enhanced bootstrap method.28 With approximately the final multivariable model. Sixteen patients (10 cases; 6
270 cases in each CD cohort, a 1:2 ratio of case:control, a 10% controls) experienced primary neuraxial block failure before

XXX 2017 Volume XXX Number XXX www.anesthesia-analgesia.org


3
Copyright 2017 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Postpartum Hemorrhage and Cesarean Delivery

group = predelivery Hb 11 g/dL]), multiple pregnancies


Table 1.Characteristics of Women With
and Without Severe Postpartum Hemorrhage (aOR = 1.8; 95% CI = 1.12.9), and government-assisted insur-
Undergoing Prelabor Cesarean Delivery ance (aOR = 1.6; 95% CI = 1.02.4). Compared with women
No Severe PPH Severe PPH who underwent primary CD, women with one previous
(n = 550) (n = 269) P Value CD had a reduced odds of severe PPH (aOR = 0.4; 95% CI =
Maternal age (y) 35 (6) 34 (6) .07 0.20.9). Women aged 30 years had a reduced odds of PPH
Weight at delivery (kg)a 84 (22) 84 (22) .94 compared with women aged <30 years (aOR = 0.7; 95% CI =
Insurance .09
Private 392 (71.3%) 176 (65.4%)
0.51.0). The model had good calibration (HosmerLemeshow
Government assisted/ 158 (28.7%) 93 (34.6%) test, P = .47) and modest discrimination (AUROC= 0.75; 95%
other CI = 0.710.78). After bootstrapping, the c-index was unal-
Race/ethnicity .001 tered. In separate models, the interactions between magne-
Non-Hispanic white 241 (43.8%) 77 (28.6%) sium with preeclampsia and gestational age at delivery were
Hispanic 130 (23.6%) 82 (30.5%)
not statistically significant (data not presented).
Non-Hispanic black 28 (5.1%) 14 (5.2%)
Asian 123 (22.4%) 76 (28.3%)
Other 28 (5.1%) 20 (7.4%) Sensitivity and Subgroup Analyses
Chronic hypertension 19 (3.4%) 23 (8.6%) .002 To determine whether point estimates for modes of anes-
Gestational age at 39 [3739] 37 [3438] <.001 thesia were influenced by women who experienced pri-
delivery (wk) mary neuraxial block failure before surgery, we performed
Parity 1 [02] 1 [02] .78
Number of earlier CDs <0.001
sensitivity analyses examining women in each CD cohort
0 204 (37.1%) 139 (51.7%) whose primary mode of anesthesia was successful. For each
1 239 (43.4%) 68 (25.3%) CD cohort, point estimates for mode of anesthesia did not
2 88 (16%) 42 (15.6%) appreciably change from those in the original multivariable
 3 19 (3.4%) 20 (7.4%) models (data not presented).
Type of pregnancy <.001
Because general anesthesia may be considered more
Singleton 516 (93.8%) 210 (78.1%)
Multiple pregnancies 34 (6.2%) 59 (21.9%)
often for women with placental disorders, for the prelabor
Placenta previa 29 (5.3%) 92 (34.2%) <.001 CD cohort, we performed an additional sensitivity analysis
GDM 80 (14.6%) 59 (21.9%) .008 to assess point estimates for anesthetic techniques exclud-
Preeclampsia 32 (5.8%) 37 (13.7%) <.001 ing women with abnormal placentation (n = 51) or placenta
Previous history of D&C 98 (17.8%) 66 (24.5%) .02 previa (n = 121). On the basis of chart review, we identified
or D&E
women with abnormal placentation based on information in
Previous history of 27 (4.9%) 13 (4.8%) .96
myomectomy the obstetricians operative note or from a placental pathol-
Predelivery hemoglobin 12.2 (1.1) 11.6 (1.4) <.001 ogy report (if available). Among those without these placental
(g/dL)b disorders, we observed comparable point estimates for anes-
Time of delivery .01 thetic techniques to those in our original multivariable model
Weekday daytime 448 (81.5%) 195 (72.5%) (CSE: aOR = 3.87; 95% CI = 1.937.73; epidural: aOR = 1.7;
Weekday night time 41 (7.5%) 32 (11.9%)
95% CI = 0.39.5; general anesthesia: aOR = 47.6; 95% CI =
Weekend 61 (11.1%) 42 (15.6%)
Mode of anesthesia <.001 7.26311.9).
Spinal 379 (68.9%) 103 (38.3%) For women undergoing intrapartum CD, we compared
CSE 156 (28.4%) 117 (43.5%) PPH frequencies among those with versus without labor
Epidural 11 (2%) 19 (7.1%) arrest before full cervical dilation. Severe PPH was more
General anesthesia 4 (0.7%) 30 (11.1%) common among women with versus without labor arrest
Type of uterine incision <.001
Transverse 531 (96.6%) 211 (78.4%)
(132/203 (65%) vs 214/382 (56%), respectively; P = .04).
Other 19 (3.4%) 58 (21.6%) Similarly, we compared PPH rates between those with
Data presented as n (%); mean (SD); median [IQR].
versus without a prolonged second stage of labor; rates
Abbreviations: CD, cesarean delivery; CSE, combined spinal-epidural; D&C, were similar in both groups (36/61 [59%] vs 99/151 [66%],
dilation and curettage; D&E, dilation and evacuation; GDM, gestational respectively; P = .37).
diabetes.
In both CD cohorts, we assessed the frequency of women
a
Data missing for 1 case and 1 control.
b
Data missing for 6 cases and 10 controls. who experienced an EBL 1500 mL, intraoperative and
postoperative RBC transfusion stratified by predelivery Hb
categories (Supplemental Digital Content, Supplemental
surgery (Supplemental Digital Content, Supplemental Figure Table 2 and Table 4, http://links.lww.com/AA/B691).
2, http://links.lww.com/AA/B691). Severe PPH was more Among cases identified in each CD cohort, rates of intraop-
common among women who underwent CSE anesthesia and erative and postoperative transfusion were highest among
general anesthesia. Clinical factors independently associated women with a predelivery Hb level <10 g/dL. Finally, to
with severe PPH are presented in Table4. In the multivari- determine whether predelivery Hb influences the magni-
able model, factors independently associated with severe tude of blood loss, we performed sensitivity analyses with
PPH were general anesthesia (aOR = 5.4; 95% CI = 1.717.1), the outcome based solely on the criteria of an EBL 1500
multiple pregnancies (aOR= 3.2; 95% CI = 1.76.3), predeliv- mL. On multivariable analyses, we observed that predeliv-
ery Hb 9.9 g/dL (aOR = 3.0; 95% CI = 1.36.9), predelivery ery Hb category was not independently associated with an
Hb = 1010.9 g/dL (aOR=2.6; 95% CI = 1.44.9 [reference EBL 1500 mL in each CD cohort (data not presented).

4
www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
Copyright 2017 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Table 2.Unadjusted and Adjusted Odds Ratios for Variables Associated With Severe Postpartum
Hemorrhage During Prelabor Cesarean Delivery
Unadjusted Odds Ratio (95% CI) Adjusted Odds Ratio (95% CI)a P Valueb
Maternal age (y) c

<34 Reference Reference


3437 1.21 (0.851.74) 1.16 (0.632.14) 1.0
38 1.24 (0.881.76) 1.17 (0.642.14) 1.0
Insurance
Private Reference Reference
Government assisted/other 1.31 (0.961.79) 0.62 (0.351.07) .09
Race/ethnicity
Non-Hispanic white Reference Reference
Hispanic 1.97 (1.352.88) 2.43 (1.025.78) .04
Non-Hispanic black 1.56 (0.783.12) 2.42 (0.698.52) .49
Asian 1.93 (1.312.83) 1.9 (0.913.94) .14
Other 2.24 (1.194.19) 2.35 (0.737.58) .4
Chronic hypertension 2.61 (1.44.89) 1.59 (0.73.61) .27
Gestational age (wk)c
38 Reference Reference
3437 3.41 (2.434.78) 1.03 (0.561.87) 1.0
33 12.57 (7.1822.0) 2.39 (0.936.18) .08
Number of earlier CDs
0 Reference Reference
1 0.42 (0.290.59) 0.65 (0.341.22) .44
2 0.70 (0.461.07) 0.67 (0.291.54) 1.0
3 1.54 (0.83.0) 0.66 (0.172.55) 1.0
Type of pregnancy
Singleton Reference Reference
Multiple pregnancies 4.26 (2.716.7) 8.0 (4.2515.04) <.001
GDM 1.65 (1.142.4) 1.52 (0.922.5) .1
Placenta previa 9.34 (5.9514.66) 6.33 (3.411.76) <.001
Previous D&C or D&E 1.5 (1.052.14) 1.88 (1.173.0) .009
Predelivery hemoglobin (g/dL)
11 Reference Reference
1010.9 1.58 (1.042.4) 1.11 (0.552.24) 1.0
9.9 2.95 (1.585.52) 2.80 (1.057.5) .04
Time of CD
Weekday daytime Reference Reference
Weekday night time 1.79 (1.12.93) 0.98 (0.42.39) 1.0
Weekend 1.58 (1.032.43) 1.64 (0.833.24) .25
Mode of anesthesia
Spinal Reference Reference
CSE 2.76 (1.983.84) 3.13 (1.715.71) <.001
Epidural 4.52 (2.0210.09) 2.54 (0.5611.44) .61
General anesthesia 30.36 (12.6173.1) 22.25 (4.9599.89) <.001
Uterine incision
Transverse Reference Reference
Other 7.68 (4.4713.21) 4.12 (2.058.26) <.001
Abbreviations: CD, cesarean delivery; CSE, combined spinalepidural; D&C, dilation and curettage; D&E, dilation and evacuation; GDM, gestational diabetes.
a
The confidence intervals and P values for all pairwise comparisons presented in the multivariable model adjusted using a Bonferroni correction to account for
multiple comparisons.
b
Statistically significant associations in the multivariable model are denoted by bold text.
c
Categories were created according to the relationship between log odds of the variable with the outcome, based on cubic spline function models.

DISCUSSION CD (intrapartum CD). Future studies are needed to deter-


Using detailed clinical data abstracted from 1669 medical mine whether confounding by indication explains the associ-
records, we performed 2 case-control studies to identify risk ations between CSE and general anesthesia with severe PPH.
factors for severe PPH during prelabor CD and intrapartum Although a number of studies have sought to identify
CD. Risk factors common to each CD cohort included gen- risk factors for PPH during CD,15,16,21,31 the majority have not
eral anesthesia, predelivery anemia, and multiple pregnan-
accounted for key differences in the maternal characteristics
cies. Other risk factors identified in our analyses were also
and delivery indications between prelabor and intrapartum
well recognized: placenta previa (prelabor CD) and cho-
rioamnionitis (intrapartum CD). We identified several risk CDs.32 Only 2 studies examined risk factors across distinct
factors that were less established, such as CSE anesthesia, CD populations; elective/nonelective CD16 and planned/
nontransverse uterine incision, earlier D&C, and Hispanic emergency CD.15 Consistent with our findings, these stud-
race (prelabor CD); and gestational diabetes (GDM), young ies identified placenta previa and general anesthesia as key
maternal age, government-assisted insurance, and primary risk factors across all CD cohorts. However, neither study

XXX 2017 Volume XXX Number XXX www.anesthesia-analgesia.org


5
Copyright 2017 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Postpartum Hemorrhage and Cesarean Delivery

labor or other intrapartum characteristics as potential pre-


Table 3.Antepartum and Intrapartum
Characteristics of Women With and Without Severe dictors in a single multivariable model for severe PPH dur-
Postpartum Hemorrhage Undergoing Intrapartum ing prelabor CD. Similarly, women with placenta previa
Cesarean Delivery are unlikely to be considered for induction or trial of labor;
No Severe PPH Severe PPH therefore, it would be intuitive to consider placenta previa
Characteristic (n = 572) (n = 278) P Value as a candidate variable only for women who undergo prela-
Maternal age (y) 32 (6) 30 (7) <.001 bor CD. However, from an epidemiological standpoint, we
Weight at delivery (kg) 80 (16)a 81 (18) .44
acknowledge that future cohort studies using nuanced clini-
Insuranceb <.001
Private 362 (63.3%) 141 (50.7%) cal data are needed to investigate whether interactions truly
Government assisted/ 210 (36.7%) 137 (49.3%) exist between relevant predictors and the presence versus
other the absence of labor or induction before CD. Large study
Race/ethnicity .01 populations would also be needed to construct adequately
Non-Hispanic white 203 (35.5%) 69 (24.8%) sized derivation and validation study cohorts.
Hispanic 180 (31.5%) 116 (41.7%)
In both CD cohorts, general anesthesia had the high-
Non-Hispanic black 35 (6.1%) 13 (4.7%)
Asian 126 (22%) 64 (23%) est adjusted odds of severe PPH compared with spinal
Other 28 (4.9%) 16 (5.8%) anesthesia. Because some anesthesiologists may prefer to
Chronic hypertension 15 (2.6%) 11 (4%) .29 use general anesthesia for patients with placenta previa
Gestational age at 39 [3840] 39 [3640] .002 or abnormal placentation,33 we excluded these conditions
delivery (wk)
in our sensitivity analyses for the prelabor cohort and
Parity 0 [01] 0 [01] .72
Number of earlier CDs .004
observed that the association between general anesthesia
0 455 (79.6%) 243 (87.4%) and severe PPH persisted, albeit with wider confidence
1 90 (15.7%) 21 (7.6%) intervals. This association has been reported in other obser-
2 27 (4.7%) 14 (5%) vational and randomized studies, with varying degrees of
Type of pregnancy <.001 magnitude.15,16,21,31,3436 General anesthesia may itself directly
Singleton 547 (95.6%) 242 (87%)
influence the likelihood of severe blood loss. Volatile agents
Multiple pregnancies 25 (4.4%) 36 (13%)
GDM 60 (10.5%) 51 (18.4%) .001 have a concentration-dependent inhibitory effect in vitro
Preeclampsia 34 (5.9%) 32 (11.5%) .004 on uterine smooth muscle contractility,37,38 thereby increas-
Predelivery hemoglobin 12.3 (1.1) 11.8 (1.5) <.001 ing the likelihood of uterine atony. Volatile agents (halo-
(g/dL)c thane, sevoflurane) and induction agents (propofol) can
Type of labor .49 also inhibit platelet function in a dose-dependent man-
Spontaneous 368 (64.3%) 172 (61.9%)
Induction 204 (35.7%) 106 (38.1%)
ner.39 However, we cannot exclude the possibility of con-
Oxytocin augmentation 387 (67.7%) 189 (68%) .92 founding by indication because patients with anticipated
Chorioamnionitis 68 (11.9%) 56 (20.1%) .001 risk factors for PPH may be more likely to receive general
Magnesium infusion 41 (7.2%) 38 (13.7%) .002 anesthesia. For example, in the prelabor CD cohort, general
Cervical dilation before CD .17 anesthesia was used for 21% of women with placenta pre-
19 cm 384 (67.1%) 204 (73.4%)
via, 46% of women with abnormal placentation, and 22%
10 cm 152 (25.6%) 61 (21.9%)
Missing 36 (6.3%) 13 (4.7%)
of women with antenatal vaginal bleeding (Supplemental
Time of CD .05 Digital Content, Supplemental Table 1, http://links.lww.
Weekday daytime 168 (29.4%) 87 (31.3%) com/AA/B691). In the prelabor CD cohort, women who
Weekday night time 237 (41.4%) 92 (33.1%) underwent CSE anesthesia had a 3-fold increased odds
Weekend 167 (29.2%) 99 (35.6%) of severe PPH compared with those receiving a single-
Mode of anesthesia <.001
shot spinal. The reason for this finding is unclear. Because
Spinal 137 (24%) 40 (14.4%)
CSE 44 (7.7%) 32 (11.5%) equivalent doses of intrathecal local anesthetic drugs are
Epidural 376 (65.7%) 181 (65.1%) commonly used for spinal and CSE anesthesia, it is doubt-
General anesthesia 15 (2.6%) 25 (9%) ful that a drug effect explains this finding. Therefore, con-
Type of uterine incision .05 founding by indication is also likely, because patients with
Transverse 562 (98.2%) 267 (96%) anticipated risk factors for PPH may require longer dura-
Other 10 (1.8%) 11 (4%)
tions of surgery and receive a CSE technique. To assess the
Data presented as n (%); mean (SD); median [IQR]. potential for bias from confounding by indication for CSE
Abbreviations: CD, cesarean delivery; CSE, combined spinal-epidural; D&C,
dilation and curettage; D&E, dilation and evacuation; GDM, gestational and general anesthesia, context-specific studies are needed
diabetes; PPH, postpartum hemorrhage. to examine the relations between these modes of anesthesia
a
Data missing for 1 patient. with severe PPH.
b
Data missing for 1 case and 7 controls.
c
Data missing for 21 cases and 7 controls.
Predelivery anemia was associated with our primary
outcome, which included a composite of major blood loss
or postpartum transfusion; this association was not con-
stratified cohorts according to the presence versus the firmed in our sensitivity analysis that focused on major
absence of labor or induction before CD. blood loss only. These findings are consistent with those by
We believe that our approach to construct separate logis- Butwick et al40 who reported that predelivery anemia is an
tic models according to CD subtype is clinically justified independent risk factor for severe postpartum anemia after
for several reasons. It is not clinically intuitive to consider CD, classified by a postpartum Hb 8 g/dL. Furthermore,

6
www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
Copyright 2017 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Table 4.Unadjusted and Adjusted Odds Ratios for Variables Associated With Severe Postpartum
Hemorrhage During Intrapartum Cesarean Delivery
Unadjusted Odds Ratio (95% CI) Adjusted Odds Ratio (95% CI)a P Valueb
Maternal age (y)c
<30 Reference Reference
30 0.63 (0.470.84) 0.69 (0.481.0) .05
Insurance
Private Reference Reference
Government assisted/other 0.39 (0.320.47) 1.56 (1.02.44) .05
Race/ethnicity
Non-Hispanic white Reference Reference
Hispanic 1.9 (1.322.71) 1.23 (0.62.51) 1.0
Non-Hispanic black 1.09 (0.552.18) 0.58 (0.171.95) 1.0
Asian 1.49 (1.02.24) 1.47 (0.772.79) .93
Other 1.68 (0.863.29) 0.94 (0.312.87) 1.0
Gestational age (wk)c
40 Reference Reference
39 1.06 (0.721.58) 1.23 (0.682.22) 1.0
3538 1.32 (0.921.89) 1.25 (0.692.28) 1.0
<35 2.29 (1.423.71) 1.24 (0.453.43) 1.0
Number of earlier CDs
0 Reference Reference
1 0.44 (0.270.72) 0.43 (0.210.91) .02
2 0.97 (0.51.89) 0.49 (0.161.49) .37
Type of pregnancy
Singleton Reference Reference
Multiple pregnancies 3.25 (1.915.54) 3.25 (1.676.31) <.001
GDM 1.92 (1.282.87) 1.83 (1.152.93) .01
Preeclampsia 2.06 (1.243.41) 1.53 (0.743.21) .25
Predelivery hemoglobin (g/dL)
11 Reference Reference
1010.9 2.64 (1.664.18) 2.6 (1.384.9) .001
9.9 3.44 (1.896.24) 2.99 (1.296.92) .005
Oxytocin augmentation 1.02 (0.751.38) 0.78 (0.471.31) .35
Chorioamnionitis 1.87 (1.272.75) 1.76 (1.132.75) .01
Magnesium infusion 2.05 (1.293.27) 1.26 (0.62.64) .54
Time of CD
Weekday daytime Reference Reference
Weekday night time 0.75 (0.531.07) 0.64 (0.391.03) .08
Weekend 1.14 (0.81.63) 1.05 (0.631.73) 1.0
Mode of anesthesia
Spinal Reference Reference
CSE 2.59 (1.454.62) 2.31 (0.915.85) .11
Epidural 1.65 (1.112.44) 2.06 (0.994.31) .06
General anesthesia 5.95 (2.6913.14) 5.44 (1.7317.12) .006
Uterine incision
Transverse Reference Reference
Other 2.32 (0.975.52) 2.75 (0.958.01) .06
Abbreviations: CD, cesarean delivery; CSE, combined spinal-epidural.
a
The confidence intervals and P values for all pairwise comparisons presented in the multivariable model adjusted using a Bonferroni correction to account for
multiple comparisons.
b
Statistically significant associations in the multivariable model are denoted by bold text.
c
Categories created according to the relationship between log odds of the variable with the outcome, based on cubic spline function models.

several obstetric societies, including The Royal College maternal Hb levels before delivery limits the likelihood
of Obstetricians and Gynecologists (UK) and the French of postpartum RBC transfusion. Third, RBC transfusion
College of Gynecologists and Obstetricians, have published may be used during acute episodes of blood loss as well
PPH guidelines recommending that a Hb level above 8 g/ as for treating symptomatic postpartum anemia in a non-
dL is a therapeutic goal.41 These observations, coupled with bleeding patient. Studies are also needed to determine best
our findings, have important clinical relevance for several approaches for anemia management during and after an
reasons. First, antenatal anemia is a modifiable risk factor episode of severe PPH.
for transfusion. Second, in the nonobstetric literature, lim- We identified a number of other well-established risk fac-
iting anemia before surgery is a key component of patient tors, including multiple gestations (both cohorts), placenta
blood management.42 With antenatal anemia affecting up previa (prelabor CD), and chorioamnionitis (intrapartum
to 25% pregnant women,43 similar initiatives may be neces- CD).5,15,31,44 Other risk factors are less recognized. Hispanic
sary to promote anemia correction before delivery. Future women had a 2-fold increased odds of severe PPH com-
studies are needed to examine whether optimization of pared with whites, after prelabor CD, but not intrapartum

XXX 2017 Volume XXX Number XXX www.anesthesia-analgesia.org


7
Copyright 2017 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Postpartum Hemorrhage and Cesarean Delivery

CD. Disparities in PPH have been described by others.45 ascertain, from chart review, whether all cases of abnor-
These may due to differences in the provision of medical mal placentation were suspected during the antenatal
care as well as underlying social, biological, and genetic period. We acknowledge this as an important limitation
factors. Only 1 study has investigated whether the type of because abnormal placentation is strongly associated with
uterine incision influences PPH risk. Combs et al21 reported major postpartum hemorrhage and massive transfusion.52
that a classical uterine incision was an independent risk However, placenta previa and number of earlier CDs are
factor. Compared with a low transverse incision, a greater well-known risk factors for abnormal placentation,53 and
degree of blood loss may occur with nontransverse inci- both variables were considered as candidate variables in
sions because thicker and more vascular myometrial tissue our regression model for the prelabor cesarean cohort.
may be surgically incised. The risk of PPH among women The findings from our analyses cannot be used to deter-
with earlier D&E or D&C may be related to altered placen- mine individual patient prediction of severe PPH. Such
tation.46 Future studies are needed to determine whether analyses would require large population-wide cohort stud-
this association is observed among women undergoing ies using nuanced clinical data. However, our data could be
intrapartum CD. The relationship between GDM and severe used to guide investigators in the selection of relevant can-
PPH may be explained by fetal macrosomia, a recognized didate variables in studies investigating prediction models
risk factor for PPH.47 In the intrapartum CD cohort, women of PPH during CD. Relevant prediction models require
with a history of 1 previous CD had a lower adjusted odds external validation in independent cohorts.
of severe PPH compared with women with no previous In conclusion, our findings suggest that the risk factor
CDs. Because of concern of uterine rupture, it is possible profile for severe PPH differs according to whether CD is
that obstetric providers are less likely to consider long peri- performed before or after the onset of labor or induction. To
ods of labor or augmentation for women undergoing trial of confirm these findings, population-wide studies are needed
labor after previous CD. This may partly explain why these to determine whether data should be partitioned according
women were at lower risk of severe PPH compared with to CD subtype. E
those without a history of previous CD.
Our study had a number of limitations. Although we ACKNOWLEDGMENTS
considered a large number of candidate variables, unmea- The authors are grateful for the assistance of Dr. G. Hilton
sured factors may have influenced PPH risk, such as and L. Tian who assisted with some of the data collection
obstetric or anesthesia provider experience. In a system- for this project. They also wish to acknowledge the sup-
atic review, Bauer et al48 previously reported that the risk port of researchers based in the Stanford Center for Clinical
of failed conversion of labor epidural analgesia to CD Informatics (Stanford School of Medicine) for identifying
anesthesia is higher if care is provided by a nonobstetric patients study from STRIDE and electronic hospital data-
anesthesiologist compared with care given by an obstetric bases. STRIDE is a research and development project at
Stanford University to create a standards-based informat-
anesthesiologist. Studies are needed to evaluate whether
ics platform supporting clinical and translational research.
specific anesthetic practices that may influence the magni- The STRIDE project is supported by the National Center for
tude of blood loss, such as the use of high concentrations Research Resources and the National Center for Advancing
of volatile anesthetic and blood product use, vary accord- Translational Sciences, National Institutes of Health,
ing to anesthesia provider experience. Because of the low through grant UL1 RR025744.
frequency of several factors, such as fibroids and prolonged
second-stage labor, we could not accurately examine asso- APPENDIX
ciations between these variables with severe PPH. We did Criteria for Arrest of First Stage of Labor and
not collect data for the frequency, timing, or indications for Prolonged Second Stage of Labor
patients who underwent intraoperative conversion from We defined arrest of active phase of the first stage of labor
neuraxial to general anesthesia. Given that combined rates if latent phase was completed and no documented evidence
of preoperative and intraoperative conversion from neurax- of cervical change for >2 hours. These criteria were also
ial to general anesthesia are low (1.7%5.8%),4951 any influ- used to define prolonged second stage for nulliparous
ence of intraoperative conversion on the point estimates for women (second stage exceeding 2 hours and 3 hours with
severe PPH among those undergoing neuraxial anesthesia and without regional anesthesia, respectively) and for
multiparous women (second stage exceeding 1 hour and 2
is likely to be small.
hours with and without regional anesthesia, respectively).
We did not assess risk factors among women who under-
These definitions were based on guidelines published by
went elective or nonelective CD because we chose to dif- the American College of Obstetricians and Gynecologists in
ferentiate our cohorts by the presence versus the absence 2003.25
of spontaneous labor or induction of labor. We chose this
approach because patient characteristics and CD indica-
DISCLOSURES
tions are known to markedly differ across these cohorts.32 Name: Alexander Butwick, MBBS, FRCA, MS.
Furthermore, determining the electiveness of CD may have Contribution: This author helped conceive the work, acquire, ana-
been more challenging, resulting in misclassification bias. lyze and interpret the data, and draft and revise the manuscript.
We identified women with abnormal placentation based Name: Bharathi Ramachandran, BS.
Contribution: This author helped acquire the data.
on information in the medical record documented after the Name: Priya Hegde, BA.
time of delivery. We elected not to include abnormal pla- Contribution: This author helped acquire the data.
centation as an a priori risk factor because we could not Name: Edward Riley, MD.

8
www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
Copyright 2017 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Contribution: This author helped conceive the work, interpret the 19. Deneux-Tharaux C, Dupont C, Colin C, et al. Multifaceted

data, and revise the manuscript. intervention to decrease the rate of severe postpartum haemor-
Name: Yasser El-Sayed, MD. rhage: the PITHAGORE6 cluster-randomised controlled trial.
Contribution: This author helped conceive the work, interpret the BJOG. 2010;117:12781287.
data, and revise the manuscript. 20. Rath WH. Postpartum hemorrhageupdate on problems

Name: Lorene M. Nelson, MS, PhD. of definitions and diagnosis. Acta Obstet Gynecol Scand.
Contribution: This author helped conceive the work, analyze and 2011;90:421428.
interpret the data, and revise the manuscript. 21. Combs CA, Murphy EL, Laros RK Jr. Factors associated with
This manuscript was handled by: Jill M. Mhyre, MD. hemorrhage in cesarean deliveries. Obstet Gynecol. 1991;77:7782.
22. Naef RW III, Washburne JF, Martin RW, Magann EF, Scanlon PH
REFERENCES Jr, Morrison JC. Hemorrhage associated with cesarean delivery:
1. Creanga AA, Berg CJ, Syverson C, Seed K, Bruce FC, Callaghan when is transfusion needed? J Perinatol. 1995;15:3235.
WM. Pregnancy-related mortality in the United States, 2006- 23. Belghiti J, Kayem G, Dupont C, Rudigoz RC, Bouvier-Colle
2010. Obstet Gynecol. 2015;125:512. MH, Deneux-Tharaux C. Oxytocin during labour and risk of
2. Grobman WA, Bailit JL, Rice MM, et al; Eunice Kennedy Shriver severe postpartum haemorrhage: a population-based, cohort-
National Institute of Child Health and Human Development nested case-control study. BMJ Open. 2011;1:e000514.
(NICHD) Maternal-Fetal Medicine Units (MFMU) Network. 24. American College of Obstetricians and Gynecologists. ACOG
Frequency of and factors associated with severe maternal mor- committee opinion no. 561: nonmedically indicated early-term
bidity. Obstet Gynecol. 2014;123:804810. deliveries. Obstet Gynecol. 2013;121:911915.
3. Wanderer JP, Leffert LR, Mhyre JM, Kuklina EV, Callaghan WM, 25. American College of Obstetrics and Gynecology Committee on
Bateman BT. Epidemiology of obstetric-related ICU admissions Practice Bulletins-Obstetrics. ACOG practice bulletin number
in Maryland: 1999-2008*. Crit Care Med. 2013;41:18441852. 49, December 2003: dystocia and augmentation of labor. Obstet
4. Al-Zirqi I, Vangen S, Forsn L, Stray-Pedersen B. Effects of Gynecol. 2003;102:14451454.
onset of labor and mode of delivery on severe postpartum hem- 26. Jones DR, Rushton L. Simultaneous inference in epidemiologi-
orrhage. Am J Obstet Gynecol. 2009;201:273.e1273.e9. cal studies. Int J Epidemiol. 1982;11:276282.
5. Bateman BT, Berman MF, Riley LE, Leffert LR. The epidemiol- 27. Hanley JA, McNeil BJ. The meaning and use of the area under
a receiver operating characteristic (ROC) curve. Radiology.
ogy of postpartum hemorrhage in a large, nationwide sample
1982;143:2936.
of deliveries. Anesth Analg. 2010;110:13681373.
28. Efron B, Tibsherani R. An Introduction to the Boostrap. New York:
6. Callaghan WM, Kuklina EV, Berg CJ. Trends in postpartum
Chapman and Hall; 1993.
hemorrhage: United States, 19942006. Am J Obstet Gynecol.
29. Combs CA, Murphy EL, Laros RK Jr. Factors associated with
2010;202:353.e1353.e6.
postpartum hemorrhage with vaginal birth. Obstet Gynecol.
7. Knight M, Callaghan WM, Berg C, et al. Trends in postpartum
1991;77:6976.
hemorrhage in high resource countries: a review and recom-
30. Sheiner E, Sarid L, Levy A, Seidman DS, Hallak M. Obstetric
mendations from the International Postpartum Hemorrhage
risk factors and outcome of pregnancies complicated with early
Collaborative Group. BMC Pregnancy Childbirth. 2009;9:55.
postpartum hemorrhage: a population-based study. J Matern
8. Reilly BM, Evans AT. Translating clinical research into clinical Fetal Neonatal Med. 2005;18:149154.
practice: impact of using prediction rules to make decisions. 31. Rossen J, Okland I, Nilsen OB, Eggeb TM. Is there an increase
Ann Intern Med. 2006;144:201209. of postpartum hemorrhage, and is severe hemorrhage associ-
9. Wasson JH, Sox HC, Neff RK, Goldman L. Clinical prediction ated with more frequent use of obstetric interventions? Acta
rules. Applications and methodological standards. N Engl J Obstet Gynecol Scand. 2010;89:12481255.
Med. 1985;313:793799. 32. Zhang J, Troendle J, Reddy UM, et al; Consortium on Safe
10. Al-Zirqi I, Vangen S, Forsen L, Stray-Pedersen B. Prevalence Labor. Contemporary cesarean delivery practice in the United
and risk factors of severe obstetric haemorrhage. BJOG. States. Am J Obstet Gynecol. 2010;203:326.e1326.e10.
2008;115:12651272. 33. Snegovskikh D, Clebone A, Norwitz E. Anesthetic manage-
11. Grotegut CA, Paglia MJ, Johnson LN, Thames B, James AH. ment of patients with placenta accreta and resuscitation strate-
Oxytocin exposure during labor among women with post- gies for associated massive hemorrhage. Curr Opin Anaesthesiol.
partum hemorrhage secondary to uterine atony. Am J Obstet 2011;24:274281.
Gynecol. 2011;204:56.e156.e6. 34. Chang CC, Wang IT, Chen YH, Lin HC. Anesthetic manage-
12. Rouse DJ, Leindecker S, Landon M, et al; National Institute ment as a risk factor for postpartum hemorrhage after cesarean
of Child Health and Human Development Maternal-Fetal deliveries. Am J Obstet Gynecol. 2011;205:462.e1462.e7.
Medicine Units Network. The MFMU Cesarean Registry: uter- 35. Rouse DJ, MacPherson C, Landon M, et al; National Institute
ine atony after primary cesarean delivery. Am J Obstet Gynecol. of Child Health and Human Development Maternal-Fetal
2005;193:10561060. Medicine Units Network. Blood transfusion and cesarean
13. Cheng YW, Delaney SS, Hopkins LM, Caughey AB. The asso- delivery. Obstet Gynecol. 2006;108:891897.
ciation between the length of first stage of labor, mode of deliv- 36. Afolabi BB, Lesi FE. Regional versus general anaesthesia for cae-
ery, and perinatal outcomes in women undergoing induction of sarean section. Cochrane Database Syst Rev. 2012;10:CD004350.
labor. Am J Obstet Gynecol. 2009;201:477.e1477.e7. 37. Yamakage M, Tsujiguchi N, Chen X, Kamada Y, Namiki A.
14. Kols T, ian P, Skjeldestad FE. Risks for peroperative exces- Sevoflurane inhibits contraction of uterine smooth muscle
sive blood loss in cesarean delivery. Acta Obstet Gynecol Scand. from pregnant rats similarly to halothane and isoflurane. Can J
2010;89:658663. Anesth. 2002;49:6266.
15. Skjeldestad FE, Oian P. Blood loss after cesarean delivery: a reg- 38. Yoo KY, Lee JC, Yoon MH, et al. The effects of volatile anesthet-
istry-based study in Norway, 19992008. Am J Obstet Gynecol. ics on spontaneous contractility of isolated human pregnant
2012;206:76.e176.e7. uterine muscle: a comparison among sevoflurane, desflurane,
16. Magann EF, Evans S, Hutchinson M, Collins R, Lanneau G, isoflurane, and halothane. Anesth Analg. 2006;103:443447.
Morrison JC. Postpartum hemorrhage after cesarean delivery: 39. Kozek-Langenecker SA. The effects of drugs used in anaesthe-
an analysis of risk factors. South Med J. 2005;98:681685. sia on platelet membrane receptors and on platelet function.
17. Dupont C, Touzet S, Colin C, et al; Groupe PITHAGORE 6. Curr Drug Targets. 2002;3:247258.
Incidence and management of postpartum haemorrhage fol- 40. Butwick AJ, Walsh EM, Kuzniewicz M, Li SX, Escobar GJ.

lowing the dissemination of guidelines in a network of 16 Patterns and predictors of severe postpartum anemia after
maternity units in France. Int J Obstet Anesth. 2009;18:320327. cesarean section. Transfusion. 2017;57:3644.
18. Waterstone M, Bewley S, Wolfe C. Incidence and predic-
41. Shaylor R, Weiniger CF, Austin N, et al. National and interna-
tors of severe obstetric morbidity: case-control study. BMJ. tional guidelines for patient blood management in obstetrics: a
2001;322:10891093. qualitative review. Anesth Analg. 2017;124:216232.

XXX 2017 Volume XXX Number XXX www.anesthesia-analgesia.org


9
Copyright 2017 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Postpartum Hemorrhage and Cesarean Delivery

42. Muoz M, Gmez-Ramrez S, Kozek-Langeneker S, et al. Fit to and meta-analysis of observational trials. Int J Obstet Anesth.
fly: overcoming barriers to preoperative haemoglobin optimi- 2012;21:294309.
zation in surgical patients. Br J Anaesth. 2015;115:1524. 49. Lee S, Lew E, Lim Y, Sia AT. Failure of augmentation of labor
43. World Health Organization. Worldwide Prevalence of Anemia epidural analgesia for intrapartum cesarean delivery: a retro-
19932005: WHO Global Database on Anemia. Geneva: WHO; 2008. spective review. Anesth Analg. 2009;108:252254.
44. Mehrabadi A, Hutcheon JA, Lee L, Kramer MS, Liston RM, 50. Palanisamy A, Mitani AA, Tsen LC. General anesthesia for
Joseph KS. Epidemiological investigation of a temporal increase cesarean delivery at a tertiary care hospital from 2000 to 2005:
in atonic postpartum haemorrhage: a population-based retro- a retrospective analysis and 10-year update. Int J Obstet Anesth.
spective cohort study. BJOG. 2013;120:853862. 2011;20:1016.
45. Bryant A, Mhyre JM, Leffert LR, Hoban RA, Yakoob MY, Bateman 51. Pan PH, Bogard TD, Owen MD. Incidence and characteristics
BT. The association of maternal race and ethnicity and the risk of of failures in obstetric neuraxial analgesia and anesthesia: a
postpartum hemorrhage. Anesth Analg. 2012;115:11271136. retrospective analysis of 19,259 deliveries. Int J Obstet Anesth.
46. Eshkoli T, Weintraub AY, Sergienko R, Sheiner E. Placenta
2004;13:227233.
accreta: risk factors, perinatal outcomes, and consequences for 52. Mhyre JM, Shilkrut A, Kuklina EV, et al. Massive blood trans-
subsequent births. Am J Obstet Gynecol. 2013;208:219.e1219.e7. fusion during hospitalization for delivery in New York State,
47. He XJ, Qin FY, Hu CL, Zhu M, Tian CQ, Li L. Is gestational 1998-2007. Obstet Gynecol. 2013;122:12881294.
diabetes mellitus an independent risk factor for macrosomia: a 53. Silver RM, Landon MB, Rouse DJ, et al; National Institute

meta-analysis? Arch Gynecol Obstet. 2015;291:729735. of Child Health and Human Development Maternal-Fetal
48. Bauer ME, Kountanis JA, Tsen LC, Greenfield ML, Mhyre
Medicine Units Network. Maternal morbidity associated
JM. Risk factors for failed conversion of labor epidural anal- with multiple repeat cesarean deliveries. Obstet Gynecol.
gesia to cesarean delivery anesthesia: a systematic review 2006;107:12261232.

10
www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
Copyright 2017 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.