fetal heart rate tracings ajog 2012-28-02

7/31/2019 Fetal Heart Rate Tracings AJOG 2012-28-02

1/5

O B S T E T R I C S

The 5-tier system of assessing fetal heart rate tracings is

superior to the 3-tier system in identifying fetal acidemiaJaclyn Coletta, MD; Elizabeth Murphy; Zachary Rubeo, MD; Cynthia Gyamfi-Bannerman, MD

OBJECTIVE: Electronic fetal heart rate monitoring was developed to

identify fetuses at risk of acidemia for intervention before adverse out-

comes. Our objective was to compare the 3-tier system with a 5-tier

system in evaluation of fetal acidemia.

STUDY DESIGN: Retrospective case-control study of patients with a fe-

tal arterial pH7 matched to the next birth that resulted in a pH7.2.

Tracings were categorized into 3- and 5-tier systems by a single re-

viewer. Sensitivities and specificities were calculated for each.

RESULTS: Twenty-four cases and 24 controls were identified. The sen-

sitivity for an orange or red tracing was higher than for category III, with

more of these in the pH7 group compared with controls (P .001).

There were significantly more green, blue, and yellow tracings in the

normal pH group compared with the pH 7 group (P .033, P

.008, P .023), respectively.

CONCLUSION: The 5-tier system had a better sensitivity than the 3-tier

system.

Key words: electronic fetal monitoring, fetal acidemia, neonatal

asphyxia, umbilical artery pH

Cite this article as: Coletta J, Murphy E, Rubeo Z, et al. The 5-tier system of assessing fetal heart rate tracings is superior to the 3-tier system in identifying fetal

academia. Am J Obstet Gynecol 2012;206:226.e1-5.

The use of intrapartum electronic fetalheart rate monitoring (EFM) is wide-spread, with 89% of singleton pregnancies

monitored in2004.1However, despitepre-

vious attempts, developing a standardizedapproach to interpretation and manage-

ment has been problematic. Although

EFM is often criticized for having a highfalse-positive rate as it is unable to predictcerebral palsy, this is not the intended

goal.2 EFM is used to detect patterns that

reflect significant hypoxia and allow inter-

vention before fetal injury.2 Initial studies

demonstrated the significance of EFM inreducing intrapartum mortality and de-

tectingfetalacidemiawhencomparedwith

intermittent auscultation but also showed

an increase in cesarean and operative vag-

inal deliveries.3,4As EFM analysis has been

limited by poor inter- and intraobserver

variability, in 2008, the National Institute

of Child HealthandHuman Development

(NICHD) along with the Society for Ma-

ternal-Fetal Medicine (SMFM) and the

American College of Obstetricians and

Gynecologists (ACOG) convened a work-shop to update definitions for fetal heart

rate (FHR) tracings and assist in the in-

terpretation and management.5,6 At this

workshop, several EFM interpretation

systems were reviewed, including the 1

by Parer and Ikeda,7 who constructed a

grid of all possible fetal heart rate pat-

terns and classified each into 1 of 5 cate-

gories using the color coding of the

Homeland Security Advisory System

green low risk, red severe risk of

academia,and/or evolutionto a more se-

rious pattern (Table 1). At the conclu-

sion of the workshop, attendees agreed

on a 3-tier interpretation system, which

is currently recommended for use in in-

trapartum fetal assessment (Table 2).5

Within the 3-tier system, category I

is characterized by a normal baseline

fetal heart rate, moderate variability,

the presence or absence of accelera-

tions, and absence of decelerations.5,8,9

This category excludes fetal acidemia.Category III tracings predict current or

impending fetal asphyxia with absentfetal heart rate variability in the pres-ence of recurrent late or variable decel-erations, bradycardia, or a sinusoidalpattern. (Table 2)5,8,9 The remainderand majority of FHR tracings that donot fit into category I or III are classi-

fied as category II, an intermediate cat-egory for which the recommendationsare less clear. Eighty percent of FHR trac-ings will fall into this category.

The primary goal of electronic fetalmonitoring is to identify hypoxemic oracidemic fetuses and intervene before anadverse outcome because persistent fetalacidemia canresult in permanent neuro-logic sequelae. Although a new systemhas been developed and its implementa-tion has been recommended, we do not

know whether this will predict acidemiabetter than existing systems as this hasnot been previously studied. Therefore,our objective was to compare the newer3-tier system with the existing 5-tier sys-tem of fetal heart rate interpretation inevaluation of fetal acidemia.

MATERIALS AND METHODS

This was a single-center, retrospectivecase-control study in which patients

with fetal acidemia were matched withnormal controls in deliveries that oc-

From Department of Obstetrics andGynecology, Columbia University MedicalCenter, New York, NY.

Received Sept. 3, 2011; revised Nov. 30,2011; accepted Dec 15, 2011.

The authors report no conflicts of interest.

Presented at the 58th annual meeting of theSociety for Gynecologic Investigation, Miami,FL, March 16-19, 2011.

Reprints not available from the authors.

0002-9378/$36.00

2012 Mosby, Inc. All rights reserved.

doi: 10.1016/j.ajog.2011.12.014

Research www.AJOG.org

226.e1 American Journal of Obstetrics &Gynecology MARCH 2012


2/5

curred between Jan. 2008 and March2010. Fetal acidemia was defined as anumbilicalartery pH of7.ThispHvaluewas chosen because although manystudies suggest that the lower limit ofnormal umbilical arterial pH may rangefrom 7.02 to 7.18, the risk of morbidity

or mortality does not increase until thepH is below 7.0.10 This pH is also one ofthe criteria fordefining hypoxic ischemicencephalopathy.10 Using our deliverydatabase, we identified all fetuses deliv-eredat34weeksgestationwithanum-bilical cord arterial pH7 where at least30 minutes of EFMwas available. Umbil-ical artery and venous pH and bloodgases are routinely measured at all deliv-eries in our institution. Results of bothsamples were available and reviewed to

ensure that arterial values were obtained.These cases were matched by plurality

and date of delivery to the next birth that

resulted in an umbilical arterial pH of

7.20, defined as a normal pH for this

study. The last 30 to 60 minutes of each

fetal heart rate tracing were printed and

analyzed as this period should best re-

flect the neonatal pH.

To decrease interobserver variability, asingle reviewer training in MFM andcer-

tified in EFM interpretation evaluated

each fetal heart rate tracing for baseline,

variability, and presence or absence of

accelerations and decelerations. This

same reviewer then categorized each

tracing into the current 3-tier classifica-

tion system (category I, II, or III) and the

5-tier system (green, blue, yellow, or-

ange, or red). The reviewer was blinded

to the patients clinical circumstancesduring the categorization process.

After the tracings were reviewed, elec-tronic medical records were used to ab-stract baseline demographic informa-tion such as age, parity, gestational age atdelivery, and mode of delivery for eachcase and control. Rates of maternal ante-partum complicationssuchaspreterm la-

bor, preterm premature ruptured mem-branes, diabetes, hypertensive disorders,or chorioamnionitis were also docu-mented and analyzed. Neonatal out-comes included birthweight, Apgarscores, umbilical artery pH, presence ofmajor anomalies, admission to the neo-natalintensivecareunit(NICU),anduseof supplemental oxygenation. Sensitivityand specificity were then calculated foreach of the NICHD and Parer categories.Student t test was used to compare the

continuous outcome variables betweengroups. This study was approved by the

TABLE 1

Summary of 5-tier FHR classification system

Variable................................................................................................................................................................................................................................................................................................................................................................................

Green: Orange:................................................................................................................................................................................................................................................................................................................................................................................

Normal baseline If moderate variability:................................................................................................................................................................................................................................................................................................................................................................................

Moderate variability -Tachycardia with severe VD, LD, or PD................................................................................................................................................................................................................................................................................................................................................................................

early or mild VD -Normal baseline with severe PD................................................................................................................................................................................................................................................................................................................................................................................

Blue: -Mild bradycardia with severe VD, LD. or PD................................................................................................................................................................................................................................................................................................................................................................................

If moderate variability: -Moderate bradycardia with severe VD, LD, or PD................................................................................................................................................................................................................................................................................................................................................................................

-Tachycardia with early or mild VD -Any severe bradycardia................................................................................................................................................................................................................................................................................................................................................................................

-Normal baseline with moderate VD or mild LD If minimal variability:................................................................................................................................................................................................................................................................................................................................................................................

If minimal variability: -Tachycardia with moderate/severe VD, mild/moderate LD, or PD................................................................................................................................................................................................................................................................................................................................................................................

-Tachycardia without decelerations -Normal baseline with moderate/severe VD, mild/moderate LD, or PD................................................................................................................................................................................................................................................................................................................................................................................

-Normal baseline early decelerations -Mild or moderate bradycardia early deceleration................................................................................................................................................................................................................................................................................................................................................................................

Yellow: If absent variability:................................................................................................................................................................................................................................................................................................................................................................................If moderate variability: -Normal baseline................................................................................................................................................................................................................................................................................................................................................................................

-Tachycardia with moderate VD, mild/moderate LD, or PD Red:................................................................................................................................................................................................................................................................................................................................................................................

-Normal baseline with severe VD, moderate/severe LD,mild/moderate PD

If minimal variability:

................................................................................................................................................................................................................................................................................................................................................................................

-Mild bradycardia early, mild/moderate VD, LD, or PD -Tachycardia with severe LD................................................................................................................................................................................................................................................................................................................................................................................

-Moderate bradycardia early decelerations -Normal baseline with severe LD or PD................................................................................................................................................................................................................................................................................................................................................................................

If minimal variability: -Mild or moderate or severe bradycardia with any VD, LD, or PD................................................................................................................................................................................................................................................................................................................................................................................

-Tachycardia with early or mild VD If absent variability:................................................................................................................................................................................................................................................................................................................................................................................

-Normal baseline with mild VD -Any baseline with any deceleration................................................................................................................................................................................................................................................................................................................................................................................

Marked variability Sinusoidal................................................................................................................................................................................................................................................................................................................................................................................FHR, fetal heart rate; LD, late deceleration; PD, prolonged deceleration; VD, variable deceleration.Adapted from Parer and Ikeda.7

Coletta.Acidemiaand FHRclassificationsystems. Am JObstet Gynecol 2012.

www.AJOG.org Obstetrics Research

MARCH 2012 American Journal of Obstetrics &Gynecology 226.e2


3/5

institutional review board at ColumbiaUniversity Medical Center.

RESULTS

There were 9680 deliveries from Jan.2008 through March2010. Of these, we

identified 30 tracings with a pH

7(0.3%). A total of 6 tracings were ex-cluded2 for gestational age less than34 weeks, 1 set of triplets and 1 mono-chorionic twin pair, and 1 for trisomy18. The last case was excluded as therewas only 15 minutes of tracing avail-able for interpretation. Therefore, ourfinal number of cases was 24.

We matched our 24 tracings with a pH7to24controlswithapH7.20.Base-line characteristics and obstetric compli-

cations were similar between the 2groups (Table 3), including gestationalage at delivery. However, there weremore cesarean deliveries in the pH 7group. Overall, the 5-tier system per-formed better than the 3-tier system inclassifying fetuses at risk of acidemia.The sensitivity and specificity for an or-

ange or redtracing was 79.2% and 100%,

respectively, with significantly more or-

ange and red tracings in the pH 7

group compared with controls (10/24,

41.7% vs 0%, P .001; 9/24, 37.5% vs

0%, P .001, respectively) (Table 4).

There were significantly more green,

blue, and yellow tracings in the normal

pH group compared with the pH 7

group (8/24, 33.3% vs 2/24, 8.3%, P

.033; 6/24, 25% vs 0/24, 0%, P .008;

10/24, 41.7% vs 3/24, 12.5%, P .023).

The 5-tier system also had a lower false-

negative rate (Table 5).

Using the 3-tier criteria, there was 1tracing in the pH 7 group that was

category I. The majority of tracings

werecategoryIIinboththepH 7(20/

24, 83%) and the control group (19/24,

79%) (Table 4). When comparing the2

groups, there was a nonsignificant

trend toward more category I tracings

in the control group (5/24, 20.8% vs

1/24, 4.2%, P .084) and more cate-

gory III tracings in the pH 7 group

(3/24, 12.5% vs 0/24, 0%, P .076)

(Table 4).

When neonatal outcomes were ana-

lyzed, both gender and birthweight

were similar between groups. As ex-

pected, neonatal morbidity was higher

in the pH7 group with a significantly

higher rate of NICU admissions and

higher rates of respiratory morbidity,

including supplemental oxygen use in

63% (15/24), continuous positive air-

way pressure (CPAP) in 50% (12/24),intubation in 13% (3/24), and respira-

tory distress syndrome (RDS) in 25%

(6/24) of cases (Table 6). There was a

higher rate of metabolic acidemia (11/

24, 46% vs 0/24, 0%, P .001) with a

mean base deficit of 12 in the pH 7

group and 3 in the pH7.20 group. In

TABLE 2

Three-tier FHR classificationsystem

Category.........................................................................................................

Category I:................................................................................................

Baseline rate: 110-160 beats perminute................................................................................................

Baseline variability: moderate................................................................................................

Late or variable decelerations: absent................................................................................................

Early decelerations: present or absent................................................................................................

Accelerations: present or absent.........................................................................................................

Category II:................................................................................................

Includes all tracings not categorizedas Category I or III

.........................................................................................................

Category III:................................................................................................

Absent baseline FHR variability andany of the following:................................................................................................

-Recurrent late decelerations................................................................................................

-Recurrent variable decelerations................................................................................................

-Bradycardia................................................................................................

Sinusoidal pattern...........................................................................................................

FHR, fetal heart rate.Adapted from Macones et al.5

Coletta.Acidemiaand FHRclassificationsystems.

Am JObstetGynecol 2012.

TABLE 3

Baseline characteristics and obstetric complications

Characteristic pH 7.2 Pvalue

Maternal age 31.4 31.3 .94..............................................................................................................................................................................................................................................

Nulliparous 15/24 (63%) 6/24 (25%) .09..............................................................................................................................................................................................................................................

Mean GA at delivery 38.7 38.8 .85..............................................................................................................................................................................................................................................

Antenatal compsa 12/24 (46%) 8/24 (33%) .25..............................................................................................................................................................................................................................................

Chorioamnionitis 2/24 (8%) 0/24 (0%) .15..............................................................................................................................................................................................................................................

Cesarean delivery 16/24 (67%) 4/24 (17%) .001..............................................................................................................................................................................................................................................

GA, gestational age.

a Included preterm labor, preterm premature ruptured membranes, diabetes, and hypertensive disorders.


TABLE 4

Categorization of FHR tracings

Category pH 7.2, % Pvalue

Green 8.3 33.3 .033..............................................................................................................................................................................................................................................

Blue 0 25.0 .008..............................................................................................................................................................................................................................................

Yellow 12.5 41.7 .023..............................................................................................................................................................................................................................................

Orange 41.7 0 .001..............................................................................................................................................................................................................................................

Red 37.5 0 .001..............................................................................................................................................................................................................................................

Category I 4.2 20.8 .084..............................................................................................................................................................................................................................................

Category II 83.3 79.2 .719..............................................................................................................................................................................................................................................

Category III 12.5 0 .076..............................................................................................................................................................................................................................................

FHR, fetal heart rate.


Research Obstetrics www.AJOG.org



4/5

addition, the mean 5-minute Apgarscore was significantly lower in the pH7 group (7.8 vs 9, P .001).

COMMENT

Principal findings of the study

When comparing the 2 classification sys-tems, we found that the 5-tier systemperformed better than the 3-tier systemin identifying fetuses at risk foracidemia.For a tracing categorized as either orangeor red, there was a 79% sensitivity and100%specificityforapH7withnofalse-positive results. Although our group re-cently published a study comparing thedifferent FHR classification systems andfound that 3-tier and 5-tier systems weresimilar in detecting very normal andvery abnormal tracings, the study didnot specifically evaluate fetal acidemia orneonatal outcome.11 To our knowledge,this is the first study in characterizng fetalpH.

3-tier vs 5-tier FHR classification

systems of EFM

The overall goal of the 3-tier system wasto make more specific recommendationsfor FHR pattern classification and intra-partum management. Its main criticismis the broad definitions within categoryII, inhibiting precise interpretation of fe-tal status, subsequent management, andultimate neonatal outcome. Twenty of24 tracings in the pH7groupand19of

24 tracings in the pH7.20 group wereclassified as category II, thus this cate-gory did not characterize fetal acidemiacompared with nonacidemic fetuses,based on our definitions. The purpose ofthe 5-tier system was to specifically clas-sify FHR patterns according to risk of fe-tal acidemia and risk of evolution to aworsening pattern, aiding obstetric

management. By using the 5-tier system,79%of fetal acidemia was correctly char-acterized by the presence of an orange or

red tracing, compared with only 12%with a category IIItracing.The 5-tier sys-tem includes more specific definitions,ultimately further delineating the trac-ings that would be included in categoryII. All tracings in the pH 7.20 group

were categorized as either blue, green, oryellow, thus ruling out neonatal aci-demia. Furthermore, no neonates in thisgroup were admitted to the NICU. Sim-ilarly, the 5-tier system better identifiedfetal tracings that resulted in lower neo-natal Apgar scores, admission to the in-tensive care unit, and subsequent oxygensupplementation. Although all fetuseswith a category III tracing were admittedto the NICU, this category only identi-fied 3 of the 24 cases, because of the very

strict criteria for category III, requiringeithera sinusoidal pattern or absent vari-ability as well as the presence of a brady-cardia, late, or variable decelerations.This differs from a red tracing as bothabsent or minimal variability with decel-erations will qualify.

It is clear that a category III fetal heartrate tracing is abnormal and conveys anincreased risk for fetal acidemia at thetime of observation. However, 83% ofthe tracings that resulted in significant

acidemia were category II, with noclearly defined management strategy. Al-though evidence of metabolic acidemiais only one of the criteria necessary toform the diagnosis of hypoxic ischemicencephalopathy, and the majority of ne-onates born with isolated acidemia donot meet the diagnosis for cerebral palsy,the ability to more accurately identify fe-tuses at risk would help decrease seriousmorbidity.

Strengths and limitations of the studyThe main strength of this study is that itis the first to directly compare the cur-rent 3-tier classifications with a previ-ously published system in characterizingfetal acidemia. The single reviewer wastrained by a workshop participant, andanalyzed each tracing before any knowl-edge of clinical history or outcome,avoiding the potential for bias. There areseveral limitations to this study. Thesmall sample size is reflective of the low

frequency of fetal acidemia and is thereason for the case-control study design.

TABLE 5

Statistical comparisons

Variable Sensitivity, % Specificity, % False negative, % False positive, %

Category I 20.8 95.8 76.0 ..............................................................................................................................................................................................................................................

Green 33.3 91.7 66.6 ..............................................................................................................................................................................................................................................

Blue 25 100 75.0 ..............................................................................................................................................................................................................................................

Green or blue 58.3 91.7 41.7 ..............................................................................................................................................................................................................................................

Category III 12.5 100 0..............................................................................................................................................................................................................................................

Orange 41.7 100 0..............................................................................................................................................................................................................................................

Red 37.5 100 0..............................................................................................................................................................................................................................................

Orange or red 79.2 100 0..............................................................................................................................................................................................................................................


TABLE 6

Neonatal outcomesVariable pH7.2 Pvalue

Mean birthweight 3303 3261 .80..............................................................................................................................................................................................................................................

Mean Apgar 1 min 5.3 8.6 .001..............................................................................................................................................................................................................................................

Mean Apgar 5 min 7.8 9 .80..............................................................................................................................................................................................................................................

Mean base deficit 12 3 .001..............................................................................................................................................................................................................................................

NICU Admission 16/24 (67%) 0/24 (0%) .001..............................................................................................................................................................................................................................................

CPAP/Intubated 14/24 (58%) 0/24 (0%) .001..............................................................................................................................................................................................................................................

RDS 6/24 (25%) 0/24 (0%) .001..............................................................................................................................................................................................................................................

CPAP, continuous positive airway pressure; NICU, neonatal intensive care unit; RDS, respiratory distress syndrome

Coletta.Acidemiaand FHRclassificationsystems. AmJObstet Gynecol 2012.

www.AJOG.org Obstetrics Research

MARCH 2012 American Journal of Obstetrics &Gynecology 226.e4


5/5

It is possible that the 3-tier system couldhave performed better with a larger sam-ple size; however, it is remarkable thatthe 5-tier system was able to differentiateacidemic from normal fetuses using thesame number of tracings. Larger datasets

maybe neededto corroborate these find-ings. In addition, it is possible that thesingle reviewer could have misclassifiedthe tracings, but the decision to haveonly 1 reviewer eliminatedthe possibilityfor interobserver variability and made itmore likely that the categories from the 2systems were concordant with the trac-ings (ie, green category I and red

category III). Finally, it is difficult tocomment on whether these findingswould be generalizable. The inter- and

intraobserver variability of the 3-tier sys-tem have been described.12 It would beinteresting to assess whether similarfindings could be obtained by other re-viewers or the same reviewer on a differ-ent occasion,but this is beyondthe scopeof this study.

CONCLUSION

The 5-tier classification system per-formed better than the 3-tier classifica-tion system in characterizing fetuses

with acidemia who required subsequent

admission to the NICU and respiratorysupport. Although the 5-tier system maybe cumbersome because of its complex-ity, it may be a worthy undertaking tobetter characterize abnormal tracingsthat subsequently result in fetal aci-

demia. Further research is needed to as-sess whether subcharacterizing the sec-ond category of the 3-tier system wouldproduce similar results. f

REFERENCES

1. Chen HY, Chauhan SP, Ananth CV, Vintzi-leos AM, Abuhamad AZ. Electronic fetal heartrate monitoring and its relationship to neonataland infant mortality in the United States. Am JObstet Gynecol 2011;204:491.e1-10.2. Freeman RK, Nageotte MP. Comments on

American College of Obstetricians and Gyne-cologists practice bulletin no. 106. Am J ObstetGynecol 2010;202:411-2.3. Vinzileos AM, Antsaklis A, Varvarigos I, Pa-pas C, Sofatzis I, Montgomery JT. A random-ized trialof intrapartum electronic fetalheart ratemonitoring versus intermittentauscultation. Ob-stet Gynecol 1993;81:899-907.4. Vintzileos AM, Nochimson DJ, Guzman ER,Knuppel RA, Lake M, Schifrin BS. Intrapartumelectronic fetal heart rate monitoring versus in-termittent auscultation: a meta-analysis. ObstetGynecol 1995;85:149-55.5. Macones GA,HankinsGD, Spong CY,HauthJ, Moore T. The 2008 National Institute of ChildHealth and Human Development workshop re-

port on electronic fetal monitoring:update on def-

initions, interpretation, and research guidelines.

Obstet Gynecol 2008;112:661-6.6. Chauhan SP, Klauser CK, Woodring TC,

Sanderson M, Magann EF, Morrison JC. Intra-partum nonreassuring fetal heart rate tracing

and prediction of adverse outcomes: interob-server variability. Am J Obstet Gynecol 2008;

199:623.e1-5.7. Parer JT, Ikeda T. A framework for standard-ized management of intrapartum fetal heart rate

patterns. Am J Obstet Gynecol 2007;197:26.e1-6.

8. American College of Obstetricians and Gyne-cologists. ACOG practice bulletin no. 106: intra-

partum fetal heart rate monitoring: nomenclature,interpretation, and general management princi-

ples. Washington, DC: The College; 2009.9. American College of Obstetricians and Gyne-

cologists.ACOG practice bulletin no. 116:man-agement of intrapartum fetal heartrate tracings.

Washington, DC: The College; 2010.10. MacLennan A. A template for defining a

causal relationship between acute intapartumevents and cerebral palsy: internationalconsen-sus statement. BMJ 1999;319:1054-9.

11. Gyamfi-Bannerman C, Grobman WA, An-toniewicz L, Hutchison M, Blackwell SC. As-

sessment of the concordance among 2-tier,3-tier, and 5-tier fetal heart rate classificationsystems. Am J Obstet Gynecol 2011;205:

288.e1-4.12. Blackwell SC, Grobman WA, Antoniewicz

L, Hutchinson M, Gyamfi-Bannerman C. In-terobserver and intraobserver reliability of the

NICHD 3-tier fetal heart rate interpretationsystem. Am J Obstet Gynecol 2011;205:

378.e1-5.

Research Obstetrics www.AJOG.org


fetal heart rate tracings ajog 2012-28-02

Documents