Background
The few studies that have addressed the relationship between severity of intrapartum fever and neonatal and maternal morbidity have had mixed results. The impact of the duration between reaching maximum intrapartum temperature and delivery on neonatal outcomes remains unknown.
Objective
To test the association of severity of intrapartum fever and duration from reaching maximum temperature to delivery with neonatal and maternal morbidity.
Study Design
This was a secondary analysis of a prospective cohort of term, singleton patients admitted for induction of labor or spontaneous labor who had intrapartum fever (≥38°C). Patients were divided into 3 groups according to maximum temperature during labor: afebrile (<38°C), mild fever (38°C–39°C), and severe fever (>39°C). The primary outcome was composite neonatal morbidity (umbilical artery pH <7.1, mechanical ventilation, respiratory distress, meconium aspiration with pulmonary hypertension, hypoglycemia, neonatal intensive care unit admission, and Apgar <7 at 5 minutes). Secondary outcomes were composite neonatal neurologic morbidity (hypoxic-ischemic encephalopathy, hypothermia treatment, and seizures) and composite maternal morbidity (postpartum hemorrhage, endometritis, and maternal packed red blood cell transfusion). Outcomes were compared between the maximum temperature groups using multivariable logistic regression. Cox proportional-hazards regression modeling accounted for the duration between reaching maximum intrapartum temperature and delivery.
Results
Of the 8132 patients included, 278 (3.4%) had a mild fever and 74 (0.9%) had a severe fever. The incidence of composite neonatal morbidity increased with intrapartum fever severity (afebrile 5.4% vs mild 18.0% vs severe 29.7%; P <.01). After adjusting for confounders, there were increased odds of composite neonatal morbidity with severe fever compared with mild fever (adjusted odds ratio, 1.93 [95% confidence interval, 1.07–3.48]). Severe fevers remained associated with composite neonatal morbidity compared with mild fevers after accounting for the duration between reaching maximum intrapartum temperature and delivery (adjusted hazard ratio, 2.05 [95% confidence interval, 1.23–3.43]). Composite neonatal neurologic morbidity and composite maternal morbidity were not different between patients with mild and patients with severe fevers.
Conclusion
Composite neonatal morbidity correlated with intrapartum fever severity in a potentially dose-dependent fashion. This correlation was independent of the duration from reaching maximum intrapartum temperature to delivery, suggesting that clinical management of intrapartum fever, in terms of timing or mode of delivery, should not be affected by this duration.
Introduction
There are nearly 4 million deliveries annually in the United States, with 1 in 15 of these complicated by an intrapartum maternal fever, , defined as a maternal temperature ≥38°C. Intrapartum fever can be caused by noninfectious events, such as epidural analgesia, prostaglandin E2 use, dehydration, hyperthyroidism, and excess ambient heat, or it can be caused by infections such as clinical chorioamnionitis or intraamniotic infection. Previous studies have demonstrated that epidural placement can increase the risk of developing an intrapartum fever, but data are mixed on whether epidural placement increases the risk of chorioamnionitis or other infection-mediated fevers. Regardless of etiology, intrapartum maternal fever has been linked to increased rates of neonatal morbidity, including low Apgar scores, respiratory distress, neonatal sepsis, meconium aspiration, neonatal encephalopathy, neonatal seizures, and neonatal intensive care unit (NICU) admission. Intrapartum fever is also associated with increased rates of maternal morbidity, including postpartum hemorrhage, labor dystocia, operative vaginal delivery, and cesarean delivery.
Why was this study conducted?
This study aimed to assess the relationship between intrapartum fever severity and neonatal morbidity and determine whether the interval between reaching maximum intrapartum fever and delivery affects the risk of neonatal morbidity.
Key findings
The incidence of composite neonatal morbidity increases with intrapartum fever severity. Severe fever was associated with increased risk of composite neonatal morbidity compared with mild fever, even after accounting for the duration between reaching maximum intrapartum temperature and delivery.
What does this add to what is known?
This study was a prospective analysis that demonstrated a potential dose-dependent relationship between intrapartum fever severity and composite neonatal morbidity risk. This association was not affected by how early or late in the labor course the maximum temperature occurred. These findings support that the management of intrapartum fever in terms of delivery timing should not be affected by the anticipated interval until delivery.
Previous studies have addressed whether the severity of intrapartum fever is associated with neonatal and maternal morbidity with mixed results. Some studies have demonstrated a dose-response increased rate of neonatal and maternal morbidity with rising intrapartum fever severity, whereas others found no difference in rates of neonatal morbidity outcomes. , , In addition, 2 studies have previously evaluated the effect of fever-to-delivery duration on neonatal and maternal outcomes. , Ashwal et al found that fever-to-delivery duration correlated with maternal but not neonatal morbidity. However, Hochler et al found that fever-to-delivery duration was not associated with neonatal or maternal morbidity. These previous studies looked at international populations with different patient characteristics and labor management strategies than those found in the United States.
The objective of this study was to assess the relationship between the severity of intrapartum fever and neonatal and maternal morbidity in a diverse, contemporary cohort. We hypothesized that the risk of maternal and neonatal morbidity will increase as the severity of intrapartum fever increases, and that this risk will be independent of the duration between reaching maximum intrapartum temperature (Tmax) and delivery.
Materials and Methods
This was a secondary analysis of a prospective cohort of term, singleton patients admitted for induction of labor or spontaneous labor between 2010 and 2015 at a tertiary academic center. Patients with major fetal anomalies, scheduled cesarean deliveries, and isolated postpartum fevers were excluded. This study was approved by the Washington University School of Medicine Human Research Protection Office (ID#: 202107028).
Trained research staff collected data regarding maternal demographics, antenatal information, intrapartum course details, and neonatal and maternal diagnoses. On our Labor and Delivery unit, intrapartum oral temperature is measured every 4 hours or within 1 hour after the most recent temperature measurement if a fever is detected. Elevated temperatures were confirmed for accuracy within 30 to 60 minutes. Patients were only treated with antibiotics if a clinical diagnosis of intraamniotic infection was made on the basis of American College of Obstetricians and Gynecologists recommendations. Patients were divided into 3 separate cohorts depending on their Tmax during labor: afebrile (<38°C), mild fever (38°C–39°C), and severe fever (>39°C). No institutional protocol existed regarding the treatment of intrapartum fever at the time of this study. Treatment of intrapartum fever with acetaminophen was administered at the discretion of patients’ primary providers. Ampicillin and gentamicin were administered when clinical chorioamnionitis was diagnosed. Per institutional protocols, pediatricians were present for deliveries complicated by chorioamnionitis.
The primary outcome was composite neonatal morbidity, which was assigned if at least 1 of the following criteria were present: umbilical artery pH <7.1, mechanical ventilation, neonatal respiratory distress, meconium aspiration with pulmonary hypertension, hypoglycemia, NICU admission, and APGAR <7 at 5 minutes. Secondary outcomes were composite neonatal neurologic morbidity (hypoxic-ischemic encephalopathy, hypothermia treatment, and seizures) and composite maternal morbidity (postpartum hemorrhage, defined as estimated blood loss >1000 mL, endometritis, and maternal packed red blood cell transfusion). Endometritis was differentiated from isolated postpartum fever by elevated maternal temperature in the setting of midline, lower abdominal pain with uterine tenderness. In patients with previous intrapartum fever or chorioamnionitis diagnosis, endometritis was diagnosed clinically if they had a persistent fever in the setting of uterine tenderness, foul smelling lochia, tachycardia, or leukocytosis. Our institutional criteria for initiating neonatal therapeutic hypothermia is moderate to severe hypoxic-ischemic encephalopathy at ≥36 weeks’ gestation at birth with any of the following: 10-minute Apgar score <5, prolonged resuscitation at birth, severe acidosis (pH <7.1) on cord or neonate blood gas analysis within 60 minutes of birth, or base deficit (>12 mmol/L) on cord or neonate blood gas analysis within 60 minutes of birth.
Baseline patient characteristics were compared between each Tmax group using analysis of variance, Kruskal–Wallis, or Fisher exact tests, as appropriate. The Cochran–Armitage test for trend was used to compare the proportions of patients with the primary and secondary outcomes across rising Tmax groups. Primary and secondary outcomes were compared between the 3 Tmax groups using multivariable logistic regression, adjusting for confounders.
The initial multivariable logistic regression model included epidural placement, prostaglandin use, nulliparity, prolonged labor (defined as labor duration ≥90th percentile), and prolonged rupture of membranes (>18 hours). Backward elimination reduced the number of variables in each model. Final model variables had ≥10% effect on the magnitude of the odds ratio associated with the exposure. Final model fit was assessed with the Hosmer–Lemeshow goodness-of-fit test. The covariates that remained in the final model were prolonged labor and prolonged rupture of membranes. Cox proportional-hazards regression modeling adjusted for the duration between reaching Tmax and delivery. Two-sided P values <.05 were considered statistically significant. Statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).
Results
Of the 8132 patients in the cohort, 278 (3.4%) had a mild fever and 74 (0.9%) had a severe fever; 65% of patients self-identified as Black and 7% identified as Latina. Patients with intrapartum fever were more likely to be younger and/or nulliparous, receive an epidural, prostaglandins, or mechanical cervical ripening, and have a longer duration of labor and membrane rupture than patients who were afebrile. Patients with intrapartum fever were also more likely to receive oxytocin, undergo operative delivery or cesarean delivery, and be diagnosed with chorioamnionitis ( Table 1 ). Of the patients with mild or severe fever, 42% were treated with acetaminophen, and 67% and 56% received ampicillin or gentamicin, respectively.
| Demographic | Afebrile N=7780 (reference) | Mild fever (38.0°C–39.0°C) N=278 | Severe fever (>39.0°C) N=74 | P value |
|---|---|---|---|---|
| Maternal age (y) | 25.9 (5.9) | 24.6 (6.0) | 24.5 (5.7) | .01 |
| BMI (kg/m 2 ) | 32.2 (7.3) | 32.8 (6.5) | 32.8 (7.9) | .23 |
| Obesity | 4338 (55.8) | 174 (62.6) | 43 (58.1) | .07 |
| Race | .13 | |||
| Black | 5039 (64.8) | 171 (61.5) | 42 (56.8) | |
| White | 1788 (23.0) | 68 (24.4) | 16 (21.6) | |
| Latina | 557 (7.2) | 19 (6.8) | 6 (8.1) | |
| Other | 344 (4.5) | 17 (6.1) | 8 (10.8) | |
| No race information | 52 (0.6) | 3 (1.0) | 2 (2.7) | |
| Gestational age (wk) | 39.3 (1.2) | 39.6 (1.2) | 39.6 (1.1) | <.01 |
| Nulliparity | 3123 (40.1) | 210 (75.5) | 53 (71.6) | <.01 |
| Induction of labor | 3417 (43.9) | 124 (44.6) | 34 (46.0) | .91 |
| Chronic hypertension | 383 (4.9) | 12 (4.3) | 1 (1.4) | .33 |
| Previous cesarean delivery | 680 (8.7) | 37 (13.3) | 4 (5.4) | .51 |
| Preexisting diabetes mellitus | 111 (1.4) | 5 (1.8) | 2 (2.7) | .58 |
| Gestational diabetes mellitus | 234 (3.0) | 12 (4.3) | 2 (2.7) | .45 |
| Hypertensive disorder of pregnancy | 1182 (15.2) | 52 (18.7) | 7 (9.5) | .10 |
| AROM | 4827 (58.1) | 168 (60.4) | 46 (62.2) | .86 |
| Duration of ROM (h) | 6.5 (11.2) | 12.9 (8.9) | 12.4 (6.7) | <.01 |
| ROM >18 h | 412 (5.4) | 54 (19.5) | 17 (23.0) | <.01 |
| Chorioamnionitis | 42 (0.5) | 263 (94.6) | 74 (100) | <.01 |
| Acetaminophen in labor | 403 (5.2) | 114 (41.0) | 34 (46.0) | <.01 |
| Ampicillin use in labor | 38 (0.5) | 184 (66.2) | 52 (70.3) | <.01 |
| Gentamicin use in labor | 23 (0.3) | 151 (54.3) | 46 (62.2) | <.01 |
| GBS-positive | 2182 (28.0) | 70 (25.2) | 14 (18.91) | .08 |
| Duration of first stage of labor (h) | 10.1 (7.6) | 16.1 (9.1) | 15.9 (7.4) | <.01 |
| Duration of second stage of labor (h) | 0.8 (1.0) | 1.5 (1.2) | 1.6 (1.5) | <.01 |
| Duration of active phase of labor (h) | 2.3 (2.6) | 5.1 (3.5) | 4.6 (3.4) | <.01 |
| Prolonged labor (duration >90th percentile of cohort) | 675 (8.6) | 77 (27.7) | 18 (24.3) | <.01 |
| Mode of delivery | <.01 | |||
| Vaginal delivery | 6287 (80.8) | 143 (51.4) | 38 (51.4) | |
| Operative vaginal | 365 (4.7) | 18 (6.5) | 9 (12.2) | |
| Cesarean delivery | 1128 (14.5) | 117 (42.1) | 27 (36.5) | |
| Infant birthweight (g) | 3230 (462) | 3384 (421) | 3453 (466) | <.01 |
| Infant sex | .91 | |||
| Male | 4002 (51.4) | 146 (52.5) | 37 (50) | |
| Female | 3778 (48.6) | 132 (47.5) | 37 (50) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
