Fetal Reserve Index · Interactive Explainer
Expert Review · AJOG 2023

The Fetal Reserve Index

A contextualized way to read electronic fetal monitoring — turning a subjective tracing into a quantitative 8-point score that flags a fetus at risk earlier than the ACOG Category system. Build a score, explore the risk factors, and see how it performs.

What this tool teaches

Screening, not diagnosis

Electronic fetal monitoring (EFM) is used in nearly 90% of US labors, yet much of the confusion around it comes from treating a screening test as if it were a diagnostic one. EFM does not diagnose fetal injury — it screens for risk. The Fetal Reserve Index (FRI) is built on that distinction.

>30%
US cesarean rate today, up from under 5% in the 1970s when EFM spread
~80%
of patients have Category II tracings — the ambiguous middle ground
14
refereed FRI studies with over 2,000 control patients to date
Publications used in this tool
1
Evans MI, Britt DW, Evans SM, Devoe LD. Improving the interpretation of electronic fetal monitoring: the fetal reserve index. Am J Obstet Gynecol. 2023;228(5S):S1129–S1143. doi:10.1016/j.ajog.2022.11.1275
Primary source — the Expert Review this entire tool is built from. Defines the FRI, its scoring, zones, and supporting data.
Primary · Expert Review
2
Eden RD, Evans MI, Evans SM, Schifrin BS. The "fetal reserve index": re-engineering the interpretation and responses to fetal heart rate patterns. Fetal Diagn Ther. 2018;43:90–104.
Source of the component list and the maternal / obstetrical / fetal risk-factor catalog (Tables 3 & 4).
Components & Risk Factors
3
Eden RD, Evans MI, Evans SM, Schifrin BS. Reengineering electronic fetal monitoring interpretation: using the fetal reserve index to anticipate the need for emergent operative delivery. Reprod Sci. 2018;25:487–497.
Prospective study reporting the drop in emergency deliveries from 17% to 4% with FRI management.
Emergency Delivery Data
4
Evans MI, Britt DW, Eden RD, Gallagher P, Evans SM, Schifrin BS. The fetal reserve index significantly outperforms ACOG Category system in predicting cord blood base excess and pH. Reprod Sci. 2019;26:858–863.
Head-to-head comparison of FRI against the Category system for acid-base prediction.
FRI vs Category
5
American College of Obstetricians and Gynecologists. Neonatal encephalopathy and neurologic outcome. D'Alton M, ed. Washington, DC: ACOG; 2014.
Source of the monograph criteria for labor-associated cerebral palsy (Table 1).
Guideline · Monograph
6
Macones GA, Hankins GD, Spong CY, Hauth J, Moore T. The 2008 NICHD workshop report on electronic fetal monitoring: update on definitions, interpretation, and research guidelines. Obstet Gynecol. 2008;112:661–666.
Establishes the Category I / II / III system the FRI is compared against.
Guideline · Categories
Interactive · Table 3 in action

Build a score

Each of the 8 categories scores 1 if normal, 0 if not. The total, divided by 8 and multiplied by 100, is the FRI. Toggle the tracing features and uterine activity directly; for the three risk-factor categories, tick any item to flag that category. Watch the score and zone update live.

Tracing & uterine activity
Risk-factor categories — tick any item to flag
8 of 8 normal
100
GREEN ZONE
All categories normal. Full fetal reserve.
02550100

How the points map to a percentage

Points (of 8)FRI %Zone
8 / 8100Green
7 / 887.5Green
6 / 875.0Green
5 / 862.5Green
4 / 850.0Yellow
3 / 837.5Yellow
2 / 825.0Red
1 / 812.5Red
0 / 80Red

Zones: Green >50% · Yellow 50%–26% · Red ≤25%. Source: Evans et al. AJOG 2023;228(5S):S1129–S1143.

Table 3

The 8 components

Four come straight off the tracing. One captures how hard the uterus is working. Three capture the clinical context. In version 1.0 of the FRI, all eight are weighted equally — each worth one point.

#CategoryWhat it captures
1Fetal heart rateBaseline rate within normal range
2Baseline variabilityModerate variability present
3AccelerationsPresence of accelerations
4DecelerationsAbsence of significant decelerations
5Increased uterine activityContraction frequency at or below threshold
6Maternal risk factorsNone present (see Risk Factors tab)
7Obstetrical risk factorsNone present
8Fetal risk factorsNone present

Each factor scored 1 if normal, 0 if not. Maximum 8/8 = 100%. Source: Evans et al. AJOG 2023.

Static vs dynamic

The risk factors (categories 6–8) are generally static — once a category loses its point, it usually stays lost until delivery. The four tracing features and uterine activity (categories 1–5) are dynamic — they change as labor progresses, with pushing and descent, and in response to intrauterine resuscitation. The FRI is recalculated for each 10-minute segment of monitoring.

Table 4

The risk-factor catalog

These are the maternal, obstetrical, and fetal conditions that cost a category its point. The presence of any item flags that whole category. They are organized by how they threaten the fetus — perfusion, oxygen delivery, and the safety of vaginal delivery.

Adapted from Eden et al. as presented in Evans et al. AJOG 2023, Table 4. Standard ACOG definitions are used for all tracing components.

Head-to-head data

FRI vs the Category system

In the authors' original series — 60 singleton term babies who entered labor with reassuring Category I tracings and went on to develop cerebral palsy, compared with 200 controls — the same cases were scored three ways. The sensitivity for identifying those CP cases differed sharply.

Sensitivity for cerebral palsy cases

ACOG monograph criteria28%
Category III45%
Fetal Reserve Index100%

In-sample sensitivity, original 60-case CP series vs 200 controls. The authors note FRI sensitivity will never be 100% in real-world practice. Source: Evans et al. AJOG 2023.

5 hr
Average time CP babies spent in the red zone, vs ~1 hour for controls
7.03
Average cord pH in the CP cases — only 27% were below the 7.00 monograph cutoff
17%→4%
Emergency deliveries in a prospective FRI-managed study
Table 1

ACOG monograph criteria for labor-associated CP

The retrospective standard the FRI is designed to anticipate. It tells you what likely happened — but only after the fact, and with rigid cutoffs (pH <7.00) that miss many CP cases.

I. Neonatal signs of an acute peripartum / intrapartum event
  • Apgar score <5 at 5 and 10 minutes
  • Umbilical artery pH <7.00 or base deficit ≥12 mmol/L
  • Acute brain injury on MRI or MR spectroscopy consistent with HIE
  • Multisystem organ failure consistent with HIE
II. Type and timing of contributing factors
  • A sentinel hypoxic/ischemic event around labor (e.g., uterine rupture, severe abruption, cord prolapse, amniotic fluid embolism, maternal collapse, fetal exsanguination)
  • FHR patterns consistent with an acute event (e.g., Category II ≥60 min with minimal/absent variability; Category I converting to Category III)
  • Brain-injury timing and pattern on imaging (MRI most sensitive 24–96 h of life)
III. Cerebral palsy type
  • Spastic quadriplegic or dyskinetic cerebral palsy

Source: ACOG. Neonatal Encephalopathy and Neurologic Outcome, 2014 (Table 1 in Evans et al. 2023). HIE = hypoxic-ischemic encephalopathy.

Table 2

Other augmented EFM systems

The FRI is not the first attempt to improve on visual tracing reading. Many automated and AI systems came before it — most rule-based, some using neural networks. A recurring theme: large trials repeatedly failed to show better perinatal outcomes.

SystemAuthorsDesignData analyzedYear
Oxford SonicaidDawes et alRule-basedAntepartum FHR1996
NST-ExpertAlonso-Betanzos et alRule-basedAntepartum FHR1992
INFANTINFANT CollaborativeRule-based + AIIntrapartum FHR2007
SIS-PortoAyres-de-Campos et alRule-basedIntrapartum FHR2000
PeriCALMElliott et alRule-based + AIIntrapartum FHR2010
STANOlofsson et alRule-basedIntrapartum FHR + fetal ECG2014
(neural network)Georgieva et alAIIntrapartum FHR2013

Source: Evans et al. AJOG 2023, Table 2. FHR = fetal heart rate.

The lab-test analogy

For centuries, diagnosing a heart attack was a gestalt of signs and symptoms. The ECG helped; then CPK isoenzymes and troponin turned it into a number anyone could read, with the same conclusion every time. The authors' goal is the same for EFM — a quantitative metric so that reading fetal status becomes as reproducible as reading a lab result.

Red-zone management

The "shot clock"

Reaching the red zone is not an order to deliver. It starts a clock — borrowed from basketball — that forces a structured, time-bound response before the situation deteriorates.

① A fetus enters the red zone (FRI ≤25%)

This is a call for expeditious assessment by senior staff who can evaluate the whole picture. Intrauterine resuscitation is usually the first step:

  • Stop IV oxytocin
  • Reposition the patient
  • Increase IV fluids
  • Administer oxygen by mask

② The 40-minute window opens

Version 1 of the protocol allows up to 40 minutes to get out of the red zone. The authors are explicit that 40 minutes is an arbitrary but reasonable starting point that may change as more data accrue.

③ If still red at 40 minutes

A 30-minute-to-delivery protocol begins, consistent with ACOG guidelines.

A surprising finding

The risk doesn't stop at delivery

Analyzing continuous neonatal heart-rate and acid-base data, the authors found that fetal reserve keeps falling for several minutes after birth before it recovers — and how long recovery takes tracks with the last FRI score before delivery.

~85%
of newborns have a significant tachycardia that resolves over the first hour
25%
of cases would read as Category III if the first 10 min of neonatal tracing were judged as fetal tracing
33%
of cases dip below −12 mmol/L base excess — the threshold of neurologic risk — early after birth
Figure 8 · Kaplan–Meier recovery

Time for neonatal heart rate to return to ≤160 bpm

Newborns grouped by their last FRI score before delivery. The lower the reserve at birth, the longer recovery took.

Green / Yellow
FRI .375–1.0
31
min average
71% recovered by 20 minutes
Red
FRI .125–.250
40
min average
49% recovered by 20 minutes
Crimson
FRI = 0.0
52
min average
28% recovered by 20 minutes

Log-rank χ² = 20.02, P < .001. The worst-FRI group (~5% of patients) averaged a neonatal tachycardia of 185 bpm with persistent loss of variability. Source: Evans et al. AJOG 2023, Figure 8.