Prevention of fetal brain injury in category II tracings

Abstract

Introduction

With category II fetal heart rate tracings, the preferred timing of interventions to prevent fetal hypoxic brain damage while limiting operative interventions remains unclear. We aimed to estimate fetal extracellular base deficit (BD_ecf) during labor with category II tracings to quantify the timing of potential interventions to prevent severe fetal metabolic acidemia.

Material and methods

A longitudinal study was conducted using the database of the Recurrence Prevention Committee, Japan Obstetric Compensation System for Cerebral Palsy, including infants with severe cerebral palsy born at ≥34 weeks' gestation between 2009 and 2014. Cases included those presumed to have an intrapartum onset of hypoxic–ischemic insult based on the fetal heart rate pattern evolution from reassuring to an abnormal pattern during delivery, in association with category II tracings marked by recurrent decelerations and an umbilical arterial BD_ecf ≥ 12 mEq/L. BD_ecf changes during labor were estimated based on stages of labor and the frequency/severity of fetal heart rate decelerations using the algorithm of Ross and Gala. The times from the onset of recurrent decelerations to BD_ecf 8 and 12 mEq/L (Decels‐to‐BD8, Decels‐to‐BD12) and to delivery were determined. Cases were divided into two groups (rapid and slow progression) based upon the rate of progression of acidosis from onset of decelerations to BD_ecf 12 mEq/L, determined by a finite‐mixture model.

Results

The median Decels‐to‐BD8 (28 vs. 144 min, p < 0.01) and Decels‐to‐BD12 (46 vs. 177 min, p < 0.01) times were significantly shorter in the rapid vs slow progression. In rapid progression cases, physicians' decisions to deliver the fetus occurred at ~BD_ecf 8 mEq/L, whereas the “decisions” did not occur until BD_ecf reached 12 mEq/L in slow progression cases.

Conclusions

Fetal BD_ecf reached 12 mEq/L within 1 h of recurrent fetal heart rate decelerations in the rapid progression group and within 3 h in the slow progression group. These findings suggest that cases with category II tracings marked by recurrent decelerations (i.e., slow progression) may benefit from operative intervention if persisting for longer than 2 h. In contrast, cases with sudden bradycardia (i.e., rapid progression) represent a challenge to prevent severe acidosis and hypoxic brain injury due to the limited time opportunity for emergent delivery.

Fetal base deficit analysis suggests labor intervention for slowly evolving category II tracings with recurrent decelerations over 2 h may prevent hypoxic brain damage. In contrast, cases with sudden bradycardia represent a challenge to prevent severe acidosis and brain injury.

Abbreviations

BD

base deficit

CP

cerebral palsy

Decels

(recurrent) decelerations

Del

delivery

ecf

extracellular fluid

EFM

electronic fetal monitoring

FHR

fetal heart rate

pCO₂

partial pressure of carbon dioxide

Key message.

Fetal base deficit analysis suggests labor intervention for slowly evolving category II tracings with recurrent decelerations over 2 h may prevent hypoxic brain damage. In contrast, cases with sudden bradycardia represent a challenge to prevent severe acidosis and brain injury.

1. INTRODUCTION

Electronic fetal monitoring (EFM) is a major tool for predicting fetal acid–base balance and is widely adopted in intrapartum management to attempt to prevent fetal hypoxic–ischemic brain injury. Several guidelines provide recommendations for clinical decision‐making and actions using three or five‐tiered classifications. ¹ , ² , ³ However, there remain concerns regarding the broad intermediate categories and low predictive ability for fetal acidemia, ⁴ , ⁵ , ⁶ and the efficacy of intrapartum EFM for reducing cerebral palsy (CP) remains unclear. ⁷

Clark et al. proposed an algorithm‐based approach to the management of category II tracings with recurrent severe variable (lasting >60 s and decreasing >60 bpm below baseline or reaching a nadir <60 bpm) or late decelerations, in which decisions for operative delivery were dependent in part upon normal labor progress. ⁸ As fetuses cycle between active sleep, moderate variability states, and quiet sleep states marked by minimal variability, the recommendation to perform cesarean or operative vaginal delivery after 30 min of minimal variability and “significant decelerations with ≥50% of contractions” may result in an increase in interventions. ⁹ Furthermore, questions remain as to the management of category II tracings with recurrent decelerations that do not meet the Clark et al. criteria as well as the specific duration of time for which observation can be continued.

Previous research regarding the fetal heart rate (FHR) evolution patterns of term and near‐term fetuses with severe CP revealed that one‐third of fetuses with severe CP experience an intrapartum onset of brain injury and that 16% of the slowly evolving patterns may be preventable. ¹⁰ However, it remains unclear as to the optimal timing of operative interventions during labor to prevent fetuses from severe hypoxia‐ischemia and metabolic acidosis resulting in brain damage while avoiding unnecessary interventions.

The American College of Obstetricians and Gynecologists (ACOG)/American Academy of Pediatrics (AAP) recommendations state that “Fetal umbilical artery pH <7.0, or base deficit greater than or equal to 12 mmol/L, or both, increases the probability that neonatal encephalopathy, if present, had an intrapartum hypoxic component; lesser degrees of acidemia decrease that likelihood.” ¹¹ Furthermore, newborn umbilical artery base deficit (BD) values of 12 to 20 mmol/L result in an incidence of CP of 2%–4%, while lesser degrees of BD have not been associated with the occurrence of CP or death. ¹² , ¹³ Thus, decisions for delivery intervention before the fetal base deficit (BD) reaches 12 mmol/L would be warranted for favorable neonatal outcomes. In regards to the timing of hypoxic brain insult during labor, the extracellular base deficit (BD_ecf)‐predicting algorithm proposed by Ross et al. ¹³ , ¹⁴ , ¹⁵ is a valuable objective assessment to estimate fetal BD during labor. ¹⁵ , ¹⁶ , ¹⁷ As the surrogate threshold level of acidosis for an increased risk of perinatal hypoxic brain injury is a BD 12 mEq/L, ¹² , ¹⁸ we defined BD 12 mEq/L as the “time of potential injury”. As a BD 8 mEq/L is approximately the mean plus one standard deviation of normal delivery umbilical artery BD, ¹⁹ , ²⁰ we defined BD 8 mEq/L as an optimal “time of intervention”. We utilized fetal tracings of participants with CP due to intrapartum hypoxia and estimated BD values using an algorithm for the assessment of BD (see Material and methods) throughout labor to infer the timing of potential interventions which may prevent hypoxic brain injury.

2. MATERIAL AND METHODS

2.1. Study population and data collection

This subgroup analysis of the nationwide longitudinal study focuses on the intrapartum FHR evolution patterns in infants with severe CP born after 34 weeks of gestation. ¹⁰ The design and main analysis of the longitudinal study have been previously described. ¹⁰ All data in the current study were extracted from the nationwide database of the Japan Obstetric Compensation System for Cerebral Palsy. ²¹ The database includes children born after 28 weeks of gestation in 2009 or later who were diagnosed with severe CP (Gross Motor Function Classification System ‐ expanded and revised ²² grade 3–5 equivalent paralysis) by the age of 5 years due to causes other than congenital disorders. Cases in whom an intrapartum insult was indicated based on the FHR evolution pattern analysis (reassuring FHR admission test followed by an abnormal pattern marked by category II tracings with recurrent decelerations), reviewed by three blinded obstetricians in our previous study, ¹⁰ were selected from a pool of 1069 infants with severe CP who were born at ≥34 weeks of gestation. Infants whose umbilical cord arterial BD_ecf values were <12 mEq/L or not available, or whose EFM tracing was inadequate to identify the time of the onset of recurrent decelerations and the time of BD_ecf ≥ 12 were excluded from the study. Umbilical cord abnormalities, including cord entanglement, insertion site abnormality (marginal or velamentous), abnormal twisting or coiling, Wharton's jelly deficiency, long (≥70 cm) or short (≤25 cm) cords, a single umbilical artery, and true knot, were confirmed at delivery. Intrauterine infection was defined as a febrile status combined with clinical or pathological chorioamnionitis or neonatal sepsis. Small‐for‐gestational age (SGA) was defined as a birthweight <10%iles of the normal Japanese values. Operative vaginal delivery was considered inappropriate if a traction with vacuum extractor or forceps was applied out of compliance with recommendations of the clinical guidelines (such as a prolonged maneuver (vacuum extraction more than five times, or for longer than 20 min), attempts at a fetal head station above 0 cm, or forceps for rotation >45 degrees). ²³ Uterine tachysystole was defined as more than five contractions in a 10‐min period, on average, persisting for more than 30 min. ²⁴ Placental abruption, umbilical cord prolapse, uterine rupture, maternal cardiopulmonary collapse, placenta previa, low‐lying placenta, and shoulder dystocia were diagnosed by the responsible physicians at each delivery facility.

2.2. EFM analysis

Three obstetricians (M.N., M.G.R., and S.M.) reviewed individual EFM tracings obtained during labor based on the National Institute of Child Health and Human Development guidelines. ²⁵ A reassuring pattern (category I) consisted of moderate FHR variability (amplitude range 6–25 bpm) and normal baseline rate (110–160 bpm) without recurrent decelerations. FHR decelerations were defined as recurrent if variable, late, or prolonged decelerations occurred with ≥50% of three or more uterine contractions per 10 min in at least two consecutive 10‐min windows and continued through delivery. The onset of the active phase and the second stage of labor were defined as ≥6 cm cervical dilation and full dilatation, respectively.

The fetal BD_ecf value during labor was calculated using the algorithm described by Ross et al. ¹³ , ¹⁴ , ¹⁵ (Figure 1). The time from the onset of recurrent decelerations to a BD_ecf 8 mEq/L, BD_ecf 12 mEq/L, and delivery (Decels‐to‐BD8, Decels‐to‐BD12, and Decels‐to‐Del, respectively), from BD_ecf 8 mEq/L, BD_ecf 12 mEq/L to delivery (BD8‐to‐Del, BD12‐to‐Del), and from decision for operative intervention to delivery (Decision‐to‐Del) were determined. The cases were statistically divided into rapid progression and slow progression groups according to the Decels‐to‐BD12 interval, and the clinical data were compared between the two groups. Some cases did not have records of the last portion of EFM tracings while preparing for cesarean section after bradycardia appeared and had high umbilical artery partial pressure of carbon dioxide (pCO₂) values, consistent with continued bradycardia. Therefore, we also divided the cases into two groups according to the umbilical artery pCO₂ value and compared the timing of delivery based on the presence or absence of severe bradycardia just prior to delivery. In this analysis, we utilized the normal values established by Yeomans et al. for umbilical artery pCO₂ (49 ± 8.4 mmHg). ²⁶ We eliminated cases of presumed air bubble contamination in which the pCO₂ values were less than the established mean – 2 SD (i.e., <32.2 mmHg).

FIGURE 1.

Scheme of fetal extracellular base deficit change during labor. During normal labor, in the absence of recurrent decelerations, the fetus enters labor with an extracellular base deficit (BD_ecf) of approximately 2 mEq/L and typically experiences no BD_ecf changes during the latent phase of labor. The BD_ecf increases by 1 mEq/L per 3–6 h of active labor, followed by a 1 mEq/L increase per hour during the second stage of labor. With recurrent decelerations, the BD_ecf increases by 0.2, 0.25, and 0.5 mEq/L per minute of mild (decreasing FHR by ≤30 bpm), moderate (by ≤60 bpm), and severe (by >60 bpm or baseline FHR <70 bpm) variable decelerations, respectively, and by 0.4 per minute of late decelerations. Prolonged deceleration or bradycardia (below 70 bpm) increases the BD by 0.5 mEq/L per minute. During recovery between decelerations, the BD_ecf decreases by 0.1 mEq/L per minute due to acid clearance through the placenta. BD_ecf, extracellular base deficit; FHR, fetal heart rate; h, hour(s); mEq/L, milliequivalents per liter; min, minute.

2.3. Algorithm for BD_ecf values

The fetal BD_ecf value at delivery was calculated using the umbilical cord arterial pH and pCO₂ values with the alignment nomogram of Siggaard‐Andersen ²⁷ as follows:

$${\mathrm{BD}}_{\mathrm{ecf}}=-\left(0.02786\times {\mathrm{pCO}}_2\times {10}^{\left[\mathrm{pH}-6.1\right]}+13.77\times \mathrm{pH}–124.58\right).$$

The estimated fetal BD_ecf was calculated at intervals during labor utilizing delivery umbilical artery BD values and the following principles ¹⁴ , ²⁸ , ²⁹ , ³⁰ :

• During normal labor in the absence of recurrent decelerations, the normal fetus enters labor with a BD of approximately 2 mEq/L and typically experiences no BD change during the latent phase of labor.

• As a result of the normal stress of uterine contractions, the BD in the normal fetus is assumed to increase by 1 mEq/L per 3–6 h of active labor, followed by 1 mEq/L increase per hour of the second stage.

• With recurrent decelerations, the BD value increases by 0.2, 0.25, and 0.5 mEq/L per minute of mild (FHR decrease ≤30 bpm), moderate (≤60 bpm), and severe (>60 bpm or baseline FHR <70 bpm) variable decelerations, respectively, and by 0.4 mEq/L per minute of late decelerations.

• Prolonged decelerations or persistent bradycardia (<70 bpm) increase the BD by 0.5 mEq/L per minute, equivalent to the effect of severe variable decelerations.

• During recovery between decelerations, the BD decreases by 0.1 mEq/L per minute due to acid clearance through the placenta (Figure 1).

2.4. Statistical analyses

Finite‐mixture models were used for dividing the cases according to the Decels‐to‐BD12 interval and umbilical artery pCO₂. Continuous and integer variables are presented as median and interquartile range, and categorical variables are presented as numbers and percentages. Univariate analyses were conducted between the two groups using the Wilcoxon signed‐rank sum test for continuous variables and the chi‐square test or Fisher's exact test for categorical variables. Statistical significance was set at p < 0.05. All analyses were conducted using STATA version 16.0 (Stata Corporation). This study was performed according to the STROBE guidelines. ³¹

2.5. Ethics statement

The ethics committee of the Japan Obstetric Compensation System for Cerebral Palsy approved this study protocol (no. 30‐01‐05 on December 17, 2021) and waived the requirement for informed consent due to the analysis of anonymized historical data.

3. RESULTS

Among 1069 infants with severe CP born at ≥34 weeks of gestation, 122 met the criteria for suspected intrapartum hypoxic–ischemic insult based upon the EFM assessments and an umbilical artery BD_ecf ≥ 12 mEq/L. Of these, 14 infants were excluded due to the lack of an adequate EFM tracing to calculate the BD_ecf. Therefore, 108 cases remained for analysis (Figure 2).

FIGURE 2.

Flow diagram of the study population. Among 1069 infants with severe CP born at ≥34 weeks of gestation, 122 met the criteria for suspected intrapartum hypoxic–ischemic insult based upon the EFM assessments and an umbilical artery BD_ecf ≥ 12 mEq/L. Of these, 108 cases were included in the final analysis. BD_ecf, extracellular base deficit; CP, cerebral palsy; EFM, electronic fetal monitoring; HI, hypoxia‐ischemia; mEq/L, milliequivalents per liter.

The timing of Decels‐to‐BD12 intervals were statistically distributed with two peaks resulting in two comparison groups (Figure 3). The median Decel‐to‐BD12 interval in the rapid progression (rapid, n = 69) and the slow progression group (slow, n = 39) was 46 and 177 min, respectively (p < 0.01). Rapid progression cases typically were characterized by a rapid transition from a reassuring pattern, often with brief periods of recurrent decelerations, to severe prolonged deceleration or bradycardia which persisted until delivery (Figure 4A). In contrast, slow progression cases typically exhibited a slowly evolving pattern with prolonged periods of recurrent variable or late decelerations, followed by reduced FHR variability and a subsequent decline of baseline rate (Figure 4B). The maternal and neonatal characteristics were similar between the groups, with the exceptions of a higher rate of maternal history of previous pregnancy loss (36% vs. 18%, p = 0.05), fertility treatment (10% vs. 0%, p = 0.05), and twin pregnancy (13% vs. 0%, p = 0.03) in the rapid group, respectively, and more male infants in the slow group (54% vs. 74%) (Table 1). Infants of both groups were born at a median of 39–40 weeks of gestation, and most deliveries required uterotonic agents. The perinatal risk factors for CP did not differ between the two groups. Nearly half of all patients had umbilical cord abnormalities, which was the most common risk factor in this cohort (n = 46: 43%), followed by intrauterine infection (n = 30: 28%) and placental abruption (n = 14: 13%). A total of 20% of the rapid and 28% of the slow cases developed severe CP without known risk factors. More infants in the rapid group were delivered by cesarean section (58% vs. 33%, p = 0.01), whereas meconium staining was noted more frequently in the slow group (36% vs. 59%, p = 0.02).

FIGURE 3.

Distribution of the time intervals from the onset of recurrent decelerations to extracellular base deficit of 12 mEq/L and the multimodal fitting curve. Cases were divided into rapid and slow groups using finite mixture models. The rapid group (<110 min) had a median interval of 46 min (n = 69) between the onset of recurrent decelerations to an extracellular base deficit (BD_ecf) of 12 mEq/L, while the slow group (≥110 min) had a median interval of 177 min (n = 39) between the onset of recurrent decelerations to a BD_ecf of 12 mEq/L. BD_ecf, extracellular base deficit; Decels‐to‐BD12 time, time interval from the onset of recurrent decelerations to extracellular base deficit 12 mEq/L; mEq/L, milliequivalents per liter; N, number; min, minutes.

FIGURE 4.

Examples of fetal tracing demonstrated rapid and slow progression. (A) Rapid progression cases typically were characterized by a rapid transition from a reassuring pattern, often with brief periods of recurrent decelerations, to severe prolonged deceleration or bradycardia which persisted until delivery. (B) Slow progression cases typically exhibited a slowly evolving pattern with prolonged periods of recurrent variable or late decelerations, followed by reduced FHR variability and a subsequent decline of baseline rate (EFM tracings were revised from: JCQHC FHR Monitoring Working Group. Cardiotocograms of cerebral palsy cases). EFM, electronic fetal monitoring; FHR, fetal heart rate; JCQHC, Japan Council for Quality Health Care.

TABLE 1.

Maternal and neonatal characteristics.

	Rapid progression (n = 69)	Slow progression (n = 39)	p‐value
Maternal characteristics
Age
Median (years)	32 (29–35)	30 (28–35)	0.36
Age ≥ 35 years	22 (31.9%)	11 (28.2%)	0.69
Prepregnancy BMI (kg/m2)	20.9 (19.7–22.6)	21.8 (20.1–23.5)	0.32
Overweight (BMI ≥25 kg/m2)	9 (13.0%)	6 (15.4%)	0.74
Nulliparous	47 (68.1%)	26 (66.7%)	0.88
Previous pregnancy loss	25 (36.2%)	7 (17.9%)	0.05
Fertility treatment	7 (10.1%)	0 (0%)	0.05
Ovulation or AIH	5 (7.3%)	0 (0%)	0.16
IVF	2 (2.9%)	0 (0%)	0.53
Twin pregnancy	9 (13.0%)	0 (0%)	0.03
Smoked cigarettes during pregnancy	1 (1.4%)	4 (10.3%)	0.06
Drank alcohol during pregnancy	1 (1.4%)	3 (7.7%)	0.13
Perinatal risk factors underlying CP (data includes duplicates)
Umbilical cord abnormalities (n = 46)	31 (44.9%)	15 (38.5%)	0.51
Intrauterine infection (n = 30)	20 (29.0%)	10 (25.6%)	0.71
Placental abruption (n = 14)	9 (13.0%)	5 (12.8%)	0.97
SGA (n = 12)	5 (7.3%)	7 (17.9%)	0.11
Uterine rupture (n = 8)	6 (8.7%)	2 (5.1%)	0.71
Umbilical cord prolapse (n = 6)	5 (7.2%)	1 (2.6%)	0.42
Inappropriate operative vaginal delivery (n = 5)	2 (2.9%)	3 (7.7%)	0.35
Maternal cardiopulmonary collapse (n = 4)	4 (5.8%)	0 (0%)	0.29
Uterine tachysystole associated with abnormal EFM (n = 2)	2 (2.9%)	0 (0%)	0.53
Maternal hemorrhage related to placenta previa or low‐lying placenta (n = 2)	1 (1.4%)	1 (2.6%)	1.00
Shoulder dystocia (n = 1)	1 (1.4%)	0 (0%)	1.00
No known risk factor (n = 25)	14 (20.3%)	11 (28.2%)	0.35
Delivery profile
Gestational age at delivery, weeks
Median	39.4 (38.1–40.4)	40.1 (38.4–40.9)	0.13
34 to <37 weeks	10 (14.5%)	4 (10.3%)	0.53
Use of uterotonic agent	40 (60.0%)	27 (69.2%)	0.17
Labor analgesia	6 (8.7%)	3 (7.7%)	1.00
Spontaneous vaginal delivery	8 (11.6%)	13 (33.3%)	<0.01
Operative vaginal delivery	21 (30.4%)	13 (33.3%)	0.76
Cesarean delivery	40 (58.0%)	13 (33.3%)	0.01
Cesarean delivery for abnormal EFM	38/40 (95.0%)	12/13 (92.3%)	0.72
Meconium staining	25 (36.2%)	23 (59.0%)	0.02
Umbilical artery pH	6.79 (6.65–6.92)	6.76 (6.68–6.87)	0.69
Umbilical artery pCO2 (mmHg) (n = 102)	103.3 (63.7–132.1)	92.3 (72.4–116.0)	0.43
No severe hypercarbia (pCO2 ≤ 92 mmHg) (n = 48)	60.9 (48.0–76.0)	72.4 (54.3–78.9)	0.25
Severe hypercarbia (pCO2 > 92 mmHg) (n = 54)	130.0 (109.0–146.5)	116.0 (104.0–127.0)	0.04
Umbilical artery BDblood (mEq/L)	23.1 (18.8–25.6)	23.5 (20.5–28.0)	0.39
Umbilical artery BDecf (mEq/L)	18.7 (15.8–21.4)	19.3 (16.4–21.3)	0.66
Time interval during the intrapartum period
Decels‐to‐BD8 time (min)	28.0 (18.0–46.0)	144.0 (110.0–206.0)	<0.01
Decels‐to‐BD12 time (min)	46.0 (31.0–67.0)	177.0 (136.0–230.0)	<0.01
Decels‐to‐Del time (min)	63.0 (54.0–97.0)	216.0 (180.0–285.0)	<0.01
BD8‐to‐Del time (min)	33.0 (24.0–51.0)	66.0 (37.0–86.0)	<0.01
BD12‐to‐Del time (min)	21.0 (11.0–32.0)	30.0 (15.0–52.0)	0.02
Decision‐to‐Del time (min)	31.0 (18.0–44.0)	29.0 (23.0–42.5)	0.89
Neonatal characteristics
Male	37 (53.6%)	29 (74.4%)	0.03
Birthweight (grams)
Median	3040 (2752–3246)	3140 (2552–3316)	0.20
Apgar score at 1 min	1 (1–2)	1 (0–2)	0.46
Apgar score at 5 min	3 (2–4)	3 (1–5)	0.22
Therapeutic hypothermia	42 (60.9%)	21 (53.8%)	0.48

Note: Data are presented as median (interquartile range) or number (percentage). Eight of nine twin pregnancies were dichorionic, and no pairs were included.

Abbreviations: AIH, artificial insemination by husband; BD, base deficit; ecf, extracellular fluid; BD8(12)‐to‐Del time, time interval from extracellular base deficit 8(12) mEq/L to delivery; BMI, body mass index; CP, cerebral palsy; Decels‐to‐BD8(12) time, time interval from the onset of recurrent decelerations to extracellular base deficit 8 (12) mEq/L; Decels‐to‐Del time, time interval from the onset of recurrent decelerations to delivery; Decision‐to‐Del time, time interval from decision to delivery; EFM, electronic fetal monitoring; IVF, in vitro fertilization; pCO₂, partial pressure of carbon dioxide; SGA, small‐for‐gestational age (birthweight < 10th percentile).

The magnitude of metabolic acidosis (BD_ecf = 19 mEq/L for both groups; p = 0.66) and hypercarbia (pCO₂ = 103 and 92 mmHg; p = 0.43) at birth were not significantly different between the rapid and slow groups. Prior to the time of BD_ecf 12 mEq/L, fetuses reached BD_ecf 8 mEq/L at a median of 28 min after the appearance of recurrent decelerations in rapid cases and at 144 min in slow cases. The Decels‐to‐Del time was approximately 1 h in the rapid progression cases, as compared to 3.5 h in the slow progression cases (p < 0.01) (Figure 5). The time between reaching BD_ecf 8 mEq/L or 12 mEq/L and delivery was longer in the slow cases than that in the rapid cases (BD8‐to‐Del time: 66 vs. 33 min, p < 0.01; BD12‐to‐Del time: 30 vs. 21 min, p = 0.02).

FIGURE 5.

Discrimination between the groups of rapid progression and slow progression. Rapid progression cases developed BD_ecf 8 mEq/L at a median of 28 min and BD_ecf 12 mEq/L at 46 min after the onset of recurrent decelerations. In contrast, in the slow progression group, fetal BD_ecf reached 8 and 12 mEq/L at 144 and 177 min, respectively. The Decels‐to‐Del time was approximately 1 h in the rapid progression cases, compared to 3.5 h in the slow progression cases. BD_ecf, extracellular base deficit; Decels‐to‐BD8(12) time, time interval from the onset of recurrent decelerations to extracellular base deficit 8(12) mEq/L; Decels‐to‐Del time, time interval from the onset of recurrent decelerations to delivery.

As terminal umbilical cord compression or placental abruption is associated with increased umbilical artery pCO₂, cases were also divided into two groups (“no severe hypercarbia” and “severe hypercarbia”) based upon umbilical artery pCO₂ values at delivery (≤92 and >92 mmHg, respectively) using the finite‐mixture models (Figure S1). The median pCO₂ values were 64 and 122 mmHg in the no severe hypercarbia and severe hypercarbia groups, respectively (p < 0.01). However, there was no significant difference in the BD12‐to‐Del time between patients with no severe hypercarbia and severe hypercarbia (rapid progression: 26 vs. 20 min, p = 0.16; slow progression: 42 vs. 25 min, p = 0.59, respectively) (Table 2).

TABLE 2.

Delivery delay from the timing of the fetal extracellular base deficit reaching 12 mEq/L, stratified by the umbilical artery partial pressure of carbon dioxide at delivery.

	Time interval (min)	p‐value
Rapid group (<110 min: n = 64)
No severe hypercarbia (pCO2 ≤ 92 mmHg) (n = 29)	26.0 (11.0–44.0)	0.16
Severe hypercarbia (pCO2 > 92 mmHg) (n = 35)	20.0 (11.0–26.0)
Slow group (≥110 min: n = 38)
No severe hypercarbia (pCO2 ≤ 92 mmHg) (n = 19)	42.0 (12.0–56.0)	0.59
Severe hypercarbia (pCO2 > 92 mmHg) (n = 19)	25.0 (22.0–43.0)

Note: Data are presented as median (interquartile range).

Abbreviations: pCO₂, partial pressure of carbon dioxide.

4. DISCUSSION

The results of the current study assess the progression of metabolic acidosis potentially resulting in CP in fetuses with category II tracings. The comparison of groups with rapid vs slow progression of acidosis suggests opportunities for clinical interventions to improve the outcomes in selected cases. Fetal BD_ecf reached 12 mEq/L within 1 h after recurrent decelerations appeared in the rapid progression group as compared to 3 h in the slow progression group. Accordingly, fetuses reached BD_ecf 8 mEq/L at 28 min after recurrent decelerations appeared in the rapid cases and at 144 min in the slow cases. These findings suggest the need for prompt recognition and decisive action in rapid cases though slow cases likely have sufficient time for attempts at in utero resuscitation or anticipated delivery.

In the current study, rapid progression cases developed metabolic acidosis levels of BD_ecf 8 mEq/L at a median of 28 min and BD_ecf 12 mEq/L at 46 min after the onset of recurrent decelerations, ultimately resulting in the development of severe CP. Importantly, physician decisions for delivery occurred 10 min before BD_ecf 12 mEq/L on average, at approximately at the time of BD_ecf 8 mEq/L. These findings suggest that although rapid Progression cases require more rapid decision‐making and prompt delivery if resuscitative maneuvers are unsuccessful, preventing hypoxic brain injury in some cases may be challenging, even with immediate decisions. Kamoshita et al. evaluated term infants with sudden fetal bradycardia and reported that long‐term neurological outcomes were favorable if the fetus was delivered by cesarean section within 25 min after bradycardia occurred. ³² Similarly, Leung et al. ³³ reported the potential risk for neonatal morbidities such as intubation, perinatal asphyxia (umbilical artery pH <7.0 with seizures and multiorgan dysfunction), and death in the cases with uterine rupture to be 18 min or longer from the onset of bradycardia. These authors stated that these poor outcomes occurred more rapidly (after 10 min) if there were preceding decelerations. Thus, a potential preexisting level of acidosis greater than that normally expected during labor ¹⁴ would shorten the time for progressive acidosis to reach threshold levels of BD 12 mmol/L. Furthermore, evidence of chronic hypoxia (i.e., fetal growth restriction) or intrauterine infection/inflammatory status may accelerate the rate of acidosis. ³⁴ These facts may explain why a few fetuses in our cohort reached a BD_ecf of 12 mEq/L in <30 min from the onset of decelerations.

In contrast to rapid progression cases, the slow progression cases demonstrated a longer time period to respond to EFM abnormalities and to delivery before pronounced acidosis occurred. As noted above, fetal BD_ecf reached 8 and 12 mEq/L at 144 and 177 min, respectively, after recurrent decelerations appeared in the slow progression group. The finding that a median of 3 h from decelerations to the threshold for severe acidosis (BD_ecf 12 mEq/L) in the slow progression group may provide a valuable guideline for clinical practice. Category II patterns with decelerations that do not meet the Clark “significant” criteria may contribute to the failure to recognize and operatively intervene, particularly in the presence of prolonged and gradual changes in the EFM patterns, as compared to rapid patterns of fetal compromise. Consistent with our results, a recent study reported a 100% intact survival rate in fetuses delivered within 145 min of gradually evolving hypoxia. ³⁵

The median umbilical artery pCO₂ values were 103 mmHg in the rapid progression group and 92 mmHg in the slow progression groups, respectively, suggesting that acute fetal bradycardia likely occurred just prior to delivery in many cases within both groups. When separated by pCO₂ levels, the median pCO₂ of ~120 mmHg in cases of severe hypercarbia (approximately 70 mmHg above the normal range) suggests the occurrence of ~10 min of terminal bradycardia, as bradycardia results in an increase of the pCO₂ by approximately 7 mmHg/min. ³⁶ However, the timing of delivery was not different between the cases with or without severe hypercarbia in either group, indicating that clinicians likely responded to the need for operative delivery when terminal bradycardia appeared.

The present study validates both the Clark ⁸ and di Pasquo ³⁵ recommendations while providing additional guidance for management of category II tracings that may not meet the criteria within these proposed algorithms. Taken together, these findings suggest that 2 h of recurrent decelerations may represent an appropriate time limit for delivery decisions, independent of progression of labor, depending upon the response to resuscitative maneuvers and assessment of additional EFM criteria (i.e., baseline rate, variability). ³⁷

The present study quantitatively validates clinical approaches to the optimal timing of operative interventions to reduce intrapartum fetal brain damage in patients with category II EFM tracings. Clark et al. ⁸ utilized severe variable or late decelerations, which commonly prompt recognition and often obstetric intervention. In the majority of our cases of recurrent variable decelerations, the patterns were typically mild or moderate. The present results suggest that recurrent decelerations (regardless of severity) extending over 2 h can result in severe levels of acidosis and the risk of neurological injury. Thus, recurrent fetal heart rate decelerations with a slowly evolving pattern over a period of 2 h may provide a quantitative frame of reference for category II tracing. Furthermore, the algorithm of BD changes during labor potentially may be incorporated into a computer analysis EFM system as a real‐time alarm in clinical practice, helping clinicians recognize gradually evolving changes and take decisive action.

The limitations of this study include the potential selection bias due to the study design and eligibility. This was a retrospective observational study only of severe CP cases associated with intrapartum asphyxia. We did not assess cases of intrapartum asphyxia without development of CP, nor of CP without intrapartum asphyxia (eg occurring prelabor or postnatal). It is understood that patients received indicated fetal resuscitation procedures (eg intravenous fluid, oxygen, position change) during the course of labor. However, using our criteria of recurrent decelerations, these interventions either preceded the onset of these decelerations or were ineffective at this time. Furthermore, the duration of fetal compromise should be considered in combination with the nature of the underlying complications. Placental abruption is often accompanied by tachysystole that may shorten the Decels‐to‐BD12 time by reducing time for the BD clearance between decelerations. Furthermore, intrauterine infection and fetal growth restriction may result in CP by pathophysiological pathways other than acute hypoxia, ³⁸ , ³⁹ which may require alternative interventions. Moreover, should there be increases in fetal metabolic demands and oxygen requirements or placental limitations on oxygen transport or acid clearance, a fetus may not be able to withstand 2 h of recurrent decelerations. However, a more rapid development of acidosis will likely result in additional evidence of potential fetal compromise (eg unstable baseline, absent FHR variability). Of course, clinicians should respond promptly to the development of category III tracings should this occur during labor. A future study focusing on the BD change rate in individual risk groups with a larger sample size would have some advantage in identifying more specific timing of potential intervention.

5. CONCLUSION

This study demonstrates that in the presence of category II tracings, fetuses achieve a BD_ecf of 12 mEq/L, the threshold of hypoxic–ischemic injury, within 1 h in rapid progression cases and 3 h in slow progression cases after the onset of recurrent decelerations. Rapid progression cases represent a challenge for the prevention of CP even with a prompt decision to deliver, recognizing that decision to incision and decision to delivery often requires up to 30 min, particularly in community hospitals. In contrast, slow progression cases may have improved outcomes in select cases when operative interventions are performed after 2 h of recurrent decelerations. Of course, all clinical interventions should be considered in the context of individual cases, the assessments of additional criteria of EFM tracings, response to potential resuscitative maneuvers, and nature of the underlying pathophysiology. Further research including a larger sample size, cases with mild CP, and cases with severe umbilical artery acidosis (BD_ecf ≥ 12 mEq/L) that does not result in CP is required before clinical guidelines can be recommended.

AUTHOR CONTRIBUTIONS

MN, MGR, and TI designed the study. MN, MGR, and SM analyzed EFM and calculated BD_ecf values. MN, MGR, and ST performed the statistical analyses, and all authors contributed significantly to interpreting the data. MN, drafted, and MGR and TI critically reviewed the manuscript. All authors read and approved the final version of the manuscript.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information

Figure S1

Nakao M, Ross MG, Magawa S, et al. Prevention of fetal brain injury in category II tracings. Acta Obstet Gynecol Scand. 2023;102:1730‐1740. doi: 10.1111/aogs.14675