Background
Fetal acidemia at the time of a scheduled cesarean delivery is generally unexpected. In the setting of reassuring preoperative monitoring, the duration of fetal acidemia in this scenario is presumably brief. The neonatal sequelae and risks associated with brief fetal acidemia in this setting are unknown.
Objective
We aimed to assess whether fetal acidemia at the time of a scheduled prelabor cesarean delivery is associated with adverse neonatal outcomes.
Study Design
This was a retrospective cohort study of singleton, term, nonanomalous, liveborn neonates delivered by scheduled cesarean delivery that was performed under regional anesthesia from 2004 to 2014 at a single tertiary care center with a universal umbilical cord gas policy. Neonates born to laboring gravidas and those whose cesarean delivery was performed for nonreassuring fetal status were excluded. All included patients had reassuring preoperative fetal monitoring. The primary outcome was a composite adverse neonatal outcome that included neonatal death, encephalopathy, therapeutic hypothermia, seizures, intubation, and respiratory distress. This outcome was compared between patients with and those without fetal acidemia (umbilical artery pH <7.2). A multivariable logistic regression was used to adjust for confounders. Cases of fetal acidemia were further characterized as respiratory, metabolic, or mixed acidemia based on additional umbilical cord gas values. Secondary analyses examining the association between the type of acidemia and neonatal outcomes were also performed.
Results
Of 2081 neonates delivered via scheduled cesarean delivery, 252 (12.1%) had fetal acidemia at the time of delivery. Acidemia was more common in breech neonates and in neonates born to gravidas with obesity and gestational diabetes mellitus. Compared with fetuses with normal umbilical artery pH, those with fetal acidemia were at a significantly increased risk for adverse neonatal outcome (adjusted relative risk, 2.95; 95% confidence interval, 2.03–4.12). This increased risk was similar regardless of the type of acidemia.
Conclusion
Even a brief period of mild acidemia is associated with adverse neonatal outcomes at the time of a scheduled cesarean delivery despite reassuring preoperative monitoring. Addressing modifiable intraoperative factors that may contribute to fetal acidemia at the time of a scheduled cesarean delivery, such as maternal hypotension and prolonged operative time, is an important priority to potentially decrease neonatal morbidity in full-term gestations.
Why was this study conducted?
Fetal acidemia is most commonly associated with labor; however, it can still be seen in up to 15% of scheduled cesarean deliveries despite reassuring preoperative fetal monitoring. We aimed to evaluate the association of this brief period of acidemia with neonatal outcomes.
Key findings
Despite reassuring preoperative fetal monitoring, mild fetal acidemia as measured using umbilical cord gases collected at the time of a scheduled cesarean delivery is associated with an increased risk for adverse neonatal outcomes. This risk is the same regardless of the type of acidemia.
What does this add to what is known?
We demonstrated an almost 3-fold increased risk for neonatal morbidity among unlabored fetuses with mild respiratory acidemia, refuting the traditional perspective that mild respiratory acidemia is inconsequential to neonatal outcomes.
Introduction
Unexpected neonatal morbidity and mortality occur in 2% to 15% of term deliveries. , Fetal acidemia is caused by intrauterine hypoxia and can predict these poor neonatal outcomes. Fetal acidemia is seen in 5% to up to 30% of term deliveries depending on how acidemia is defined. An assessment of fetal umbilical artery (UA) blood gases at the time of delivery via umbilical cord sampling is recommended by the American College of Obstetricians and Gynecologists (ACOG) and the American Academy of Pediatrics for cases in which the fetal metabolic status is in question. Identification of neonates with acidemia at the time of delivery then allows for additional pediatric surveillance and interventions to decrease long-term adverse outcomes.
Normal UA pH is reported to be 7.24 to 7.28, but the degree of acidemia associated with neonatal morbidity is commonly debated. , Historically, a UA pH of <7.0 was used as a predictor of neonatal morbidity. However, recent data suggest that mild acidemia, with a UA pH of 7.0 to 7.2, also predicts poor neonatal outcomes in a dose-dependent fashion. , The type of acidemia has also been considered in the risk determination of neonatal sequelae with the thought that respiratory acidemia, a consequence of brief impaired gas exchange, does not increase the risk for neonatal morbidity. , , Metabolic and mixed respiratory and metabolic acidemia, however, reflect prolonged hypoxia with a shift to anaerobic metabolism and confer an increased risk for poor neonatal outcomes.
Most cases of fetal acidemia are seen following labor as a consequence of interruptions in placental gas exchange during contractions but can also occur in the absence of impaired gas exchange, such as in cases of maternal diabetic ketoacidosis. , Fetal acidemia is rare at the time of a scheduled prelabor cesarean delivery (CD). , Many hypotheses exist to explain acidemia at the time of prelabor CD, including poor uterine and placental perfusion caused by anesthesia-associated hypotension or compression of the inferior vena cava by the gravid uterus in the supine position or placental disruption.
The association of acidemia at the time of prelabor CD with neonatal outcomes is not well described. We examined the incidence of fetal acidemia at the time of a scheduled CD and tested the association between acidemia and adverse neonatal outcomes. We hypothesized that fetal acidemia at the time of scheduled prelabor CD would not be associated with neonatal morbidity given the brief nature of its course.
Materials and Methods
This was a retrospective cohort study of all singleton neonates delivered via scheduled CD under regional anesthesia (spinal or epidural) at a single tertiary care center between June 2004 and December 2014. Neonates were excluded if delivery was performed before 37 weeks’ gestation based on best obstetrical assessment, major anomalies were diagnosed before or after delivery, CD was preceded by labor or rupture of membranes, CD was performed or expedited for nonreassuring fetal status, or validated umbilical cord gases were unavailable. The Human Research Protection Office at the Washington University School of Medicine approved this study before data collection and analysis (201811176).
In accordance with the universal institutional umbilical cord gas policy, a segment of the umbilical cord was double clamped after delivery of the neonate. Delayed cord clamping was not performed during the study period. UA and venous samples were extracted by trained medical personnel and sent for blood gas analysis. Whole blood was analyzed centrally to measure the UA pH, pCO 2 , and pO 2 . In 2010, UA base excess and lactate were added to the standard umbilical cord analysis. Umbilical cord blood gas measurements were validated by including only samples that included both arterial and venous samples and if arterial pH values were at least 0.02 less than venous samples to exclude double sampling of the UA.
Maternal demographic, health, pregnancy, and delivery details and neonatal outcomes were retrospectively collected from the charts by a practicing obstetrical clinician or trained perinatal research nurse. Macrosomia was defined as an infant with a birthweight ≥4000 g and small for gestational age (SGA) was defined as birthweight <10th percentile for gestational age by sex-specific birthweight curves. At our institution, all fetuses of cases scheduled for a CD undergo a nonstress test at the time of admission and have continuous fetal heart rate monitoring in the operating room during the regional anesthesia procedure and until anesthesia is gauged to be adequate for CD. This monitoring was reviewed by a practicing obstetrical clinician, masked to the UA pH and neonatal outcome, and was categorized based on the ACOG-recommended criteria. Fetal monitoring was considered discontinuous if the fetal heart rate was not measured for >1 continuous minute in the operating room. Neonates were excluded from the cohort if the CD was converted to urgent or emergent status for nonreassuring fetal status in the operating room. Maternal intraoperative hypotension was defined as a maternal systolic blood pressure (SBP) of <90 mm Hg or a >30% decrease in SBP from admission SBP to SBP before delivery of the neonate. During the study period, a liter bolus of intravenous fluid before proceeding to the operating room was the standard of care, prophylactic vasopressor use was not employed, and choice and dose of vasopressor to treat hypotension were determined by the anesthetic provider, either an anesthesiology resident, certified registered nurse anesthetist, or anesthesiologist.
The primary exposure was fetal acidemia, defined as a UA pH <7.2. The primary outcome was a composite adverse neonatal outcome that included neonatal death, encephalopathy, therapeutic hypothermia, seizures, intubation, and respiratory distress. Neonatal encephalopathy was defined using the ACOG and American Academy of Pediatrics definitions. Therapeutic hypothermia was based on institution-specific indications for treatment (ie, neurologic examination demonstrating moderate to severe encephalopathy and either a UA pH <7.1, base deficit >12 mmol/L, Apgar score <5 at 10 minutes of life, or evidence of acute perinatal event). Respiratory distress was diagnosed by the presence of persistent nasal flaring, subcostal or intercostal retractions, grunting requiring intervention or need for intubation, supplemental O 2 , or positive pressure to maintain O 2 saturations >95%. A neonatal nurse trained in resuscitation was present for all scheduled CDs. The resuscitation team alerted the in-house neonatologist in the setting of an abnormal neonatal examination. Umbilical cord gas results were not available during the initial resuscitation, but critical results were called to the supervising neonatology provider when available. These results prompted serial examination for encephalopathy, but a diagnosis of encephalopathy required an abnormal neurologic examination. Results were otherwise available in the gravida’s chart after CD. Each neonate was included in the composite once if it experienced any of the included complications. The composite outcome was compared between neonates with and those without fetal acidemia. A sensitivity analysis was performed to assess the risk for adverse neonatal outcomes at additional UA pH cutoffs of 7.1 and 7.0.
To investigate the association between the type of acidemia and neonatal outcomes, fetal acidemia was classified as respiratory, metabolic, mixed, or other in the neonates born after the addition of lactate and base excess measurements to umbilical cord blood gas analysis. Respiratory acidemia was defined as a UA pH <7.2 with UA pCO 2 >60 mm Hg and normal UA lactate and base excess. Metabolic acidemia was defined as UA pH <7.2 with UA lactate >6 mmol/L or UA base deficit ≥12 mmol/L and normal UA pCO 2 . Mixed acidemia was defined as a UA pH <7.2 with UA pCO 2 >60 mm Hg and UA lactate >6 mmol/L or UA base deficit ≥12 mmol/L. The remaining cases (UA pH <7.2 with normal UA pCO 2 and lactate) were categorized as other.
Student t tests, Wilcoxon rank-sum tests, chi-squared tests, and Fisher exact tests were used as appropriate. A P value of <.05 was considered significant. Multivariable logistic regression was used to estimate the association between fetal acidemia and the composite adverse neonatal outcome. After identifying candidate confounders by statistical difference in univariate analysis and biologic plausibility, backward stepwise elimination was performed, eliminating factors with <10% impact on the model. The final model controlled for maternal obesity and time from anesthesia induction to delivery. Goodness-of-fit of the final model was tested using the Hosmer-Lemeshow test. The Zhang method was used to approximate relative risk (RR) and adjusted RR (aRR) from the odds ratio (OR) and adjusted OR. All analyses were performed using Stata version 12.1 (StataCorp LLC, College Station, TX).
Results
Of the 2081 neonates born who met inclusion criteria, fetal acidemia was present in 252 cases (12.1%). Acidemia was more common in breech neonates and those born to gravidas with obesity or gestational diabetes mellitus ( Table 1 ). The incidence of intraoperative maternal hypotension, use of vasopressor therapy, and increased times from anesthesia induction and skin incision to delivery were all more common in cases with acidemia. Gestational age at delivery, birthweight, continuity and category of fetal heart rate tracing, and type of regional anesthesia were similar between cases with and those without fetal acidemia.
| Characteristics | Fetal acidemia (n=252) | No fetal acidemia (n=1829) | P value |
|---|---|---|---|
| Fetal or neonatal factors | |||
| Gestational age at delivery (wk) | 39 (38–39) | 39 (38–39) | .64 |
| Birthweight (g) | 3342.5 (3075–3780) | 3350 (3050–3685.4) | .08 |
| Small for gestational age | 17 (6.7) | 131 (7.2) | .81 |
| Macrosomia | 36 (14.3) | 193 (10.6) | .08 |
| Breech presentation | 43 (17.1) a | 177 (9.7) a | <.01 a |
| Male sex | 122 (48.4) | 931 (50.9) | .46 |
| Nuchal cord | 36 (14.3) | 303 (16.6) | .36 |
| Umbilical cord knot | 1 (0.4) | 19 (1.0) | .50 |
| Discontinuous fetal heart rate monitoring | 45 (33.6) | 360 (34.7) | .80 |
| Category 2 fetal heart rate monitoring | 16 (12.3) | 108 (10.7) | .58 |
| Maternal and surgical factors | |||
| Age (y) | 28 (24.5–32) | 28 (24–33) | .80 |
| BMI (kg/m 2 ) | 35.9 (30.2–44.9) a | 31.4 (26.5–38.2) a | <.01 a |
| Obesity | |||
| Nonobese | 59 (24.0) a | 740 (42.5) a | <.01 a |
| Class 1 obesity | 55 (22.4) | 411 (23.6) | .67 |
| Class 2 obesity | 36 (14.6) | 251 (14.4) | .93 |
| Class 3 obesity | 102 (40.4) a | 427 (23.4) a | <.01 a |
| Gestational weight gain (lbs) | 30 (20–42) | 30 (20–41) | .76 |
| Nulliparity | 30 (11.9) | 173 (9.5) | .22 |
| Indication for CD | <.01 a | ||
| Elective repeat | 186 (78.7) | 1439 (73.8) | |
| Breech presentation | 43 (17.1) | 177 (9.7) | |
| Other | 231 (11.7) | 23 (9.1) | |
| Previous CD | 200 (79) | 1498 (81.9) | .33 |
| >1 previous CD | 83 (32.9) | 623 (34.1) | .72 |
| Chronic hypertension | 22 (8.8) | 137 (7.5) | .49 |
| Hypertensive disorder of pregnancy | 20 (7.9) | 117 (6.4) | .36 |
| Preexisting diabetes | 7 (2.8) | 32 (1.8) | .26 |
| Gestational diabetes | 22 (8.8) a | 99 (5.4) a | .04 a |
| Smoking | 52 (20.6) | 343 (18.8) | .48 |
| Alcohol exposure | 5 (2.0) | 18 (1.0) | .19 |
| Marijuana use | 14 (5.6) | 136 (7.6) | .27 |
| Other illicit drug use | 3 (1.2) | 40 (2.2) | .30 |
| Spinal anesthesia | 236 (93.6) | 1725 (94.3) | .67 |
| Epidural anesthesia | 138 (54.8) | 984 (53.8) | .77 |
| Maternal intraoperative hypotension | 159 (64.1) a | 846 (46.8) a | <.01 a |
| Use of vasopressors before delivery | 202 (82.1) a | 1295 (72.3) a | <.01 a |
| Time from anesthesia induction to delivery (min) | 33 (26–42) a | 29 (24–37) a | <.01 a |
| Time from skin incision to delivery (min) | 13 (8–20) a | 10 (7–14) a | <.01 a |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
