Characterization of Uterine Response to Misoprostol based on

Electrohysterogram

C. Benalcazar Parra

, R. Monfort-Orti

, Y. Ye-Lin

, J. Alberola-Rubio

, A. Perales Marin

J. Mas-Cabo

, J. Garcia-Casado

and G. Prats-Boluda

Centro de Investigación e Innovación en Bioingeniería, Universidad Politécnica de Valencia, Valencia, Spain

Obstetric service, Hospital Universitario y Politécnico La Fe de Valencia, Valencia, Spain

Keywords: Labor Induction, Electrohysterogram, EHG, Misoprostol, Cervical Ripening.

Abstract: When the maternal and fetal risks of prolonging gestation are higher than the benefits, labor induction is

performed in order to stimulate uterine contractions and to facilitate cervical ripening. Nevertheless, not all

cases end up in successful induction leading to an increase in the rate of caesarean sections. The aim of this

study was to study the electrophysiological uterine response to misoprostol drug by obtaining and analyzing

the evolution of temporal and spectral parameters from uterine electromyogram (electrohysterogram, EHG)

records picked up during the first 4 hours of labor induction. Successful inductions showed a progressive

increase in amplitude and a frequency shift in spectral content towards higher frequencies approximately

120 min after the initiation of labor induction; such response was not seen in failed inductions. In

conclusion, the electrophysiological response caused by effect of misoprostol in pregnant women has been

characterized by EHG parameters which showed patterns in their evolution that were different for

successful and failed labor inductions. EHG recording and analysis could serve as a very helpful tool to

predict the success of labor induction and hence reduce risks and facilitate labor management in this

frequent clinical situation.

1 INTRODUCTION

Labor induction is a common practice in obstetrics

accounting for 20% of all births (Martin, 2010). It is

indicated when the maternal and fetal risks of

prolonging gestation are higher than the benefits. It

is performed in order to stimulate uterine

contractions and to facilitate cervical ripening.

However almost 20% of all inductions are associated

to caesarean deliveries (Seyb, 1999). Successful

induction has been defined as woman achieving

active stage of labor within 48 h, although other

studies as Gudex et al (Gudex, 1993) define it as

vaginal delivery at any moment after the induction

of labor. A key factor to reduce the maternal and

fetal distress, time and hospital costs is to know

whether a woman will reach active stage of labor or

not within a few hours of the onset of labor

induction.

Assessment of cervical condition is important to

estimate successful of labor induction. The common

method to assess cervical status and to predict labor

induction is by Bishop score (Bishop, 1964). Despite

Bishop score is widely used by clinicians, some

studies have reported that this score is subjective and

with poor reproducibility (Faltin-Traub, 2004).

Other studies as Watson et al (Watson, 1996) and

Crane et al (Crane, 2004) have proven that the

information of cervical length, maternal age, height,

weight, parity, and birth weight could be used for

predicting labor induction success. Despite all these

efforts, no reliable models are available for

predicting success of labor induction in the clinical

setting.

The characterization of uterus

electrophysiological condition may provide

additional information for predicting labor induction

success. In this respect, electrohysterogram (EHG),

which is the recording of the uterine electrical

activity on abdominal surface, has been proven to be

a promising technique to assess not only the uterine

dynamics but also the uterus electrophysiological

condition (Garfield, 2007; Schlembach, 2009;

Euliano, 2013). To date, no efforts have been made

in predicting labor induction success by using EHG

records. So as to explore the possibility of predicting

labor induction success during the first 4 hours with

Benalcazar Parra C., Montfort-Orti R., Ye-Lin Y., Alberola-Rubio J., Perales Marin A., Mas-Cabo J., Garcia-Casado J. and Prats-Boluda G.

Characterization of Uterine Response to Misoprostol based on Electrohysterogram.

DOI: 10.5220/0006146700640069

In Proceedings of the 10th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2017), pages 64-69

ISBN: 978-989-758-212-7

the help of EHG information, the aim of this work

was to study the electrophysiological uterine

response to labor induction drug (misoprostol) by

obtaining and analyzing the evolution of temporal

and spectral parameters of EHG.

2 MATERIAL AND METHODS

2.1 Recording Protocol

The study adheres to the Declaration of Helsinki and

was approved by the ethics committee of the

hospital. All subjects were informed of the nature of

the study and signed an informed consent form.

Recording sessions were carried out in healthy

women with singleton pregnancies who were

determined to undergo labor induction at Hospital

Universitario y Politécnico La Fe de Valencia. All

women underwent labor induction with vaginal

insertion of 25 µg of Misoprostol (Misofar, BIAL,

Coronado, Portugal) with repeated doses every four

hours up to three administrations. The induction was

considered successful when woman reaches active

stage of labor. Women were divided into successful

(GS) and failed induction (GF) groups according to

such criterion.

For surface myoelectrical recording, the skin was

carefully prepared using an abrasive paste in order to

reduce the contact impedance. Subsequently four

monopolar Ag/AgCl electrodes were placed on

abdominal surface as shown in figure 1: 2 electrodes

supraumbilical at each side of the medial line with 8

cm of inter-electrode distance corresponding to EHG

monopolar records (M1, M2), 1 reference electrode

in the right hip and 1 ground electrode in the left hip.

The recorded abdominal signals were amplified and

band-pass filtered (0.1-30 Hz) using a commercial

biosignal amplifier (Grass 15LT+4 Grass 15A94;

Grass Instruments, West Warwick, RI) and

digitalized at a sampling rate of 1000 Hz.

Simultaneous tocographic signal (TOCO) was

obtained by placing a pressure sensor on the

abdominal surface and conditioned using monitor

Corometrix 250cx series (GE HealthCare, General

Electric Company, USA) which sends the data to a

personal computer at a sampling rate of 4 Hz. Both

signals were stored for subsequent analysis. Each

recording session comprises: 30 minutes of basal

activity (without drug) and 4 hours after the onset of

labor induction. Finally the following obstetrical

data were also collected: maternal age, gestational

age, body mass index (BMI), gestations, parity,

Bishop score, fetal weight and time to delivery from

the initiation of the labor induction.

Figure 1: Configuration of surface electrodes for obtaining

monopolar EHG recordings (M1, M2).

2.2 Signal Preprocessing and Analysis

Since EHG signal distributes its energy between 0.1

Hz and 4 Hz, a pre-processing step was required.

First, monopolar signals were band-pass filtered

between 0.1 and 4 Hz in order to remove undesired

components. Then signals were down sampled to a

sampling rate of 20 Hz to reduce the amount of data

and computational cost. Finally, bipolar EHG signal

were obtained digitally to reduce common mode

interferences as follows:

Bip = M2 - M1 (1)

Since counting the number of uterine

contractions in a period time is traditionally used to

study the progress of labor, first of all the number of

contractions detected by conventional method

(TOCO) and EHG was valued. TOCO contractions

were considered when detecting a significant rise in

amplitude (>20mmHg) and a minimum duration of

30s. EHG bursts associated to uterine contraction

were manually segmented from bipolar signal using

the following criteria: burst had to correspond to

significant increase in amplitude and/or frequency in

comparison to basal activity, present a minimum

duration of 30 seconds, and no artifact evidence

should have been observed during contraction.

Next, in order to characterize the EHG bursts, a

set of temporal and spectral parameters were

obtained in the range of (0.2-1Hz) since important

information is contained in this range (Garfield,

2007) (Marque, 1986):

• Duration (s)

• Peak to peak amplitude (μV)

• Mean frequency: MF (Hz)

• Dominant frequency in 2 frequency ranges:

DF1 (0.2-1 Hz) and DF2 (0.34-1 Hz)

• Ratio between the energy content in high

(0.34-1 Hz) and low frequency ranges (0.2-

0.34 Hz): H/L ratio

Reference

Ground

Characterization of Uterine Response to Misoprostol based on Electrohysterogram

So as to analyze the uterine electrophysiological

response to labor induction drugs, the median values

of parameters computed from EHG-bursts in

windows of 30 minutes were worked out for each

patient. Then, for each parameter and window of

analysis, the mean and standard deviation were

calculated over all the patients of each group (GS,

GF).

3 RESULTS

For a total of 14 patients, 79.6% ended in successful

induction. The obstetrics and clinical parameters of

the study groups are summarized in Table 1. Women

in both groups, GS and GF groups, were comparable

in terms of maternal age (31.9 ± 3.62 vs. 32±2) and

bishop scores (1.36±1.03 vs. 1.33±1.53). Time to

delivery was slightly shorter for women ending in

successful inductions than for those who did not

(22.98 ± 10.8 h vs. 26.62 ± 9.81 h) and newborn

weight was also shorter for GS group (3381 ± 484 g

vs. 3541.67 ± 9.81 h).

Table 1: Obstetrical and clinical parameters of patients

induced with misoprostol. Mean (std).

Variable

Success

N=11

Failure

N=3

Maternal age (years) 31.9 ± 3.62 32 ± 2

Gestational age (days) 287.64 ± 1.96 287.0 ± 0

BMI (kg/m2) 30.88 ± 6.15 31.13 ± 1.36

Gestations 1.27 ± 0.47 1 ± 0

Parity 0.18 ± 0.40 0 ± 0

Bishop score 1.36 ± 1.03 1,33 ± 1,53

Newborn weight (g) 3381 ± 484 3541.67 ± 271

Time to delivery (h) 22.98 ± 10.8 26.62 ± 9.81

Figure 2 shows the total number of contractions

detected by TOCO records and EHG records. For all

Figure 2: Contractions detected by TOCO and EHG

records.

recording session, the total number of detected

contractions in EHG records was almost double than

in TOCO records (627 vs. 324).

Figure 3 shows an EHG recording from a

woman who reached active stage of labor. When

comparing the EHG-burst obtained during the last

hour and basal activity, the EHG-burst amplitude

was significantly greater, contraction duration was

shortened, and the biopotentials oscillation was

faster suggesting the appearance of high frequency

components in EHG-burst.

Figure 3: Morphological changes in EHG burst of a

patient of the study (successful induction).

Means and standard deviations of the computed

parameters in each analysis window of 30 minutes

for GS and GF are shown in table 2 and evolution of

mean values are depicted in figure 4. For GS group,

EHG-burst duration tends to decrease, with values of

88.6±23.6 seconds at basal state (prior to drug

administration) to values of 61.6±13.0 seconds at the

last analysis window. By contrast, no clear trends

were observed in this parameter for GF group.

In regard to EHG-bursts amplitude, GS group

presented higher uterine contraction intensity than

GF group even in the basal state. In addition GS

group experiment a growing trend with values of

124.1±39.8 μV at basal state up to 167.8±58.9 μV at

the last analysis window. By contrast, the EHG-burst

amplitude changes were less noticeable in GF group

(from 83.4±15.1 μV to 102.9±11.2 μV).

Concerning to the spectral parameters, as it can

be seen in figure 4, mean frequency, DF1, DF2 and

H/L ratio parameters undergo an increasing trend for

GS group, showing a shift of the EHG-burst spectral

content towards higher frequencies (Fast wave high).

In GS patients, this shift is clearly revealed in the

H/L ratio (1.1±0.6 basal state vs. 2.8±2.0 4 hours

after labor induction onset), which measures the

energy ratio in between the high and low frequency

bands. Although similar H/L ratio values are

obtained for both GS and GF groups, this parameter

100

200

300

400

500

600

700

TOCO EHG

50 µV

10 s

30 60 90 120 150

180

210

240

270

-200

200

50 µV

10 s

t (min)

Amplitude

(µV)

Recording time after misoprostol

administration (4 h)

BIOSIGNALS 2017 - 10th International Conference on Bio-inspired Systems and Signal Processing

did not show a clear trend after labor induction onset

in the GF group, resulting in a considerably lower

value than GS group at hour 4.

4 DISCUSSION

Labor induction has become a common practice in

obstetrics. The number of labor inductions has

increased significantly in recent years. Studies

explain that this rise is due to an increase in

medically indicated induction as well as an increase

in elective induction (Grobman, 2007). Despite this,

not all inductions have successful outcomes,

resulting in an increase in the rate of caesarean

section. It is important to know whether a woman

will reach or not a successful induction so that

clinicians may be able to better plan deliveries,

preventing maternal and fetal stress which can

appear in long induction processes.

In this study, it was analyzed the response of

misoprostol drug used for preinduction cervical

ripening not only with traditional obstetrical

indicators, but also with parameters that characterize

the resulting myoelectrical activity of the uterus. Our

data shows that 25 µg of Misoprostol with repeated

dose up to three administrations has similar efficacy

in terms of success rate in comparison with other

authors (Mayer, 2016). Accordingly to Mayer et al,

success inductions with 200 µg of misoprostol

occurred in 77.3% of the cases and our results shows

a 79.6% of success inductions.

In addition, for patients involved in the study,

the time between induction and delivery for success

group was 22.98 ± 10.8 h. Although this result does

not match with those of other authors (Papanikolaou

et al, 2004) who obtain a shorter value of time to

delivery 11.9 h, this could be explained because of

the different dose employed for labor induction.

Instead of 25µg of misoprostol, they used a 50µg

dose with repeated doses up to three times.

On the other hand, our results show that EHG

records have better performance in detecting uterine

contractions than TOCO records (627 vs. 324).

Given that nowadays the number of contractions is a

basic parameter to assess labor progress, this

technique could provide a better assessment in this

sense. This is also consistent with other studies that

have seen the limitations concerning to the use of

TOCO records for monitoring uterine dynamics in

other conditions (Euliano, 2009).

Concerning to the characteristics of the EHG, the

results of this work indicate that patients from GS

and GF experiment a different electrophysiological

response to the induction drug. Failure group, except

for contraction duration, did not show any clear

trend. In contrast, in success group a gradual

increase is evident in values of EHG-burst amplitude

as well as spectral parameters. These results suggest

that misoprostol, as an agent for the stimulation of

uterine activity, acts favoring the increase of cell

junctions (gap junction); thus increasing the total

number of active cells during EHG-burst (Garfield,

2007) and so the presence of more intense

contractions and an increased ratio of the cells’

excitability. In comparison with basal state,

significant changes in EHG characteristics begin to

show as early as about 90-120 minutes after

misoprostol administration. These results coincide

with other authors who have analyzed the effect of

misoprostol on uterine contractility (Arronson et al,

2004). This study reported that the first effect of

misoprostol is an increment in uterine tonus. Then

after 1-2 h, the tonus began to decrease and is

replaced by regular uterine contractions being the

uterine activity, measured in Montevideo units,

significantly greater after 2 h of misoprostol

administration. Moreover the time required to

manifest the changes in EHG characteristics when

inducing with misoprostol is coherent with the

pharmacokinetic studies (Tang, 2002). In such study

it was found that plasma concentration of

misoprostol gradually increases after 400µg of

vaginal administration and peak plasma

concentration is reached between 75 and 80 minutes.

After that, plasma concentration slowly decreases

and undetectable levels of drug are seen even after 6

Furthermore, other authors have analyzed the

evolution of the EMG activity up to 12 hours after

administration of dinoprostone which is another

commonly drug used for labor induction (Aviram et

al, 2014). They found that uterine EMG activity,

defined as mean electrical activity of the uterine

muscle over a period of 10 minutes, did not change

significantly during the first two hours, and then

increased between 2 and 8 hours after dinoprostone

administration. Another study that analyzed the

EMG activity during labor induction with oxytocin

and dinoprostone (Toth, 2005), found statistically

significant difference in the uterine activity index

between successful and unsuccessful labor induction

after 210 minutes of labor induction onset In

comparison with our results, the uterus response to

misoprostol is faster than the response to

dinoprostone shown in those other studies. Other

authors have used DF2 parameter of EHG to analyze

Characterization of Uterine Response to Misoprostol based on Electrohysterogram

Table 2: Mean and standard deviation of parameters in each 30 minutes period. S: Successful induction F: Failure

induction.

Parameter 0' 30' 60' 90' 120' 150' 180' 210' 240'

Duration

(s)

S 88.6±23.6 87.5±25.2 76.3±16.7 87.9±25.4 76.7±20.9 70.0±13.6 69.6±16.1 60.8±6.7 61.6±13.0

88.5±20.9 97.3±11.0 95.8±33.3 83.7±27.0 89.7±45.9 75±2.8 85.8±23.7 64.0±7.0 90.8±49.6

Amplitude

(µV)

124.1±39.8 124.3±33.9 135.8±46.9 141.6±37.9 130.4±35.2 155.2±52.7 162.1±83.1 152.6±65.3 167.8±58.9

83.4±15.1 96.1±11.7 93.0±31.7 86.2±20.6 96.2±16.9 92.2±12 84.7±13.5 101.6±13.1 102.9±11.2

MF (Hz)

0.37±0.03 0.38±0.02 0.39±0.03 0.38±0.04 0.40±0.03 0.40±0.04 0.40±0.04 0.40±0.05 0.42±0.04

0.37±0.02 0.36±0.02 0.39±0.02 0.36±0.02 0.37±0.02 0.36±0.03 0.38±0.02 0.37±0.01 0.39±0.02

DF1 (Hz)

0.31±0.04 0.29±0.05 0.33±0.08 0.32±0.06 0.34±0.10 0.37±0.08 0.37±0.10 0.36±0.11 0.39±0.08

0.30±0.04 0.31±0.04 0.32±0.06 0.29±0.03 0.30±0.04 0.29±0.02 0.29±0.02 0.29±0.03 0.31±0.01

DF2 (Hz)

0.41±0.03 0.41±0.04 0.43±0.05 0.41±0.04 0.43±0.05 0.43±0.06 0.44±0.05 0.44±0.06 0.45±0.05

0.40±0.03 0.40±0.03 0.41±0.05 0.40±0.05 0.40±0.02 0.41±0.03 0.42±0.20 0.44±0.03 0.40±0.02

H/L

1.1±0.6 1.3±0.6 1.4±0.6 1.4±0.8 1.7±1.2 1.9±1.2 2.0±1.3 2.2±1.6 2.8±2.0

1.0±0.3 0.9±0.3 1.4±0.5 0.9±0.3 1.4±0.9 0.9±0.4 1.2±0.4 0.9±0.4 1.2±0.5

Figure 4: Temporal evolution of the EHG-burst parameters for women induced with misoprostol.

the effect of nifedipine which is tocolytic agent to

decrease uterine activity (Vinken, 2010). Our work

confirms the utility of spectral parameters to

characterize the electrophysiological response of the

uterus to drugs, and points to the possibility of

predicting the success of labor induction by

analyzing the EHG in the first 4 hours. For this final

aim, a more comprehensive database and the use of

machine learning methods would be needed.

5 CONCLUSION

To conclude, 25 µg of misoprostol with repeated

doses up to 3 administrations showed good

performance for labor induction. EHG can provide

relevant information about the electrophysiological

state in labor induction, In contrast to failed

inductions, a successful labor induction is associated

to remarkable changes in the characteristics of the

BIOSIGNALS 2017 - 10th International Conference on Bio-inspired Systems and Signal Processing

EHG signal (burst amplitude increase and energy

shift towards higher frequencies) after 120 minutes

of vaginal drug administration. These results point

that EHG analysis could help to predict the outcome

of labor induction.

ACKNOWLEDGEMENTS

The Ministry of Economy and Competitiveness and

the European Regional Development Fund

(DPI2015-68397-R) and Bial’s company support

this work.

REFERENCES

Martin J A, Hamilton B E, Ventura S J. et al. Births: final

data for 2010. Natl Vital Stat Rep. 2012;61:1–72.

Seyb ST. Risk of cesarean delivery with elective induction

of labor at term in nulliparous women. Obstet Gynecol

1999, 94:600–607

Gudex G. Induction of labour with prostaglandin E2 a

prospective audit. NJ Med J 1993; 106: 78–80.

Bishop EH. Pelvic scoring for elective induction. Obstet

Gynecol 1964;24:266–8

Faltin-Traub EF, Boulvain M, Faltin DL, Extermann P,

Irion O. Reliability of the Bishop score labour

induction at term. Eur J Obstet and Gynecol Reprod

Biol 2004; 112: 178–181

Watson WS, Stevens D, Welter S, Day D. Factor

predicting successful labor induction. Obstet Gynecol

1996; 88: 990–992.

Crane JM. Predictors of successful labor induction with

oral or vaginal misoprostol. J Matern Fetal Neonat

Med 2004;15: 319–323.

Garfield R, Maner W, "Physiology and electrical activity

of uterine contractions," Semin. Cell Dev. Biol., vol.

18, no. 3, pp. 289-295, June 2007

A. Aronsson, M. Bygdeman, and K. GemzellDanielsson,

“Effects of misoprostol on uterine contractility

following different routes of administration,” Hum.

Reprod., vol. 19, no. 1, pp. 81–84, Jan. 2004

Schlembach D, Maner WL, Garfield RE, Maul H.

Monitoring the progress of pregnancy and labor using

electromyography. European Journal of Obstetrics,

Gynecology, and Reproductive Biology 2009;

144(Suppl 1): S33–9.

Euliano TY, et al. Monitoring uterine activity during

labor: A comparison of 3 methods. Am J Obstet

Gynecol. 2013;208(1):66.e1–66.e6.

Marque C, Duchene JM, Leclercq S, Panczer GS,

Chaumont J. Uterine EHG processing for obstetrical

monitoring. IEEE Trans. Biomed. Eng. 1986; 33(12):

1182-7

Grobman, William A. Elective Induction: When? Ever?

Clinical obstetrics and gynecology (0009-9201). 50

(2), p. 537.

Mayer R, Oppelt P, Shebl O, Allerstofer, C, Weiss C.

Initial clinical experience with a misoprostol vaginal

insert in comparison with a dinoprostone insert for

inducing labor. Euro. Jour. of Obst. and Gyneco. and

Reprod. Biol. , Volume 200 , 89 – 93.

Papanikolaou E, Plachouras N, Drougia A, Andronikou S,

Vlachou C, Stefos T, Paraskevaidis E, Zikopoulos K.

Comparison of Misoprostol and Dinoprostone for

elective induction of labour in nulliparous women at

full term: A randomized prospective study.

Reproductive Biology and Endocrinology 2004, 2:70.

Vinken M, Rabotti C, Mischi M, van Laar J, Oei S.

Nifedipine-induced changes in the electrohysterogram

of preterm contractions: feasibility in clinical practice.

Obstet Gynecol, vol 2010, 2010, pp 1-8.

Tang OS, Schweer H, Seyberth HW, Lee SW and Ho PC

(2002) Pharmacokinetics of different routes of

administration of misoprostol. Hum Reprod 17,332-

336.

Aviram A, Melamed N, Hadar E, Raban O, Hiersch L,

Yogev Y, et al. Effect of Prostaglandin E2 on

Myometrial Electrical Activity in Women Undergoing

Induction of Labor. J Perinatol. 2014;31:413–8.

Toth T. Transcutaneous electromyography of uterus in

prediction of labor outcome induced by oxytocine and

prostaglandine shape. Gynaecologia et perinatologia:

journal for gynaecology, perinatology, reproductive

medicine and ultrasonic diagnostic, vol 14 no 2, 2005.

Characterization of Uterine Response to Misoprostol based on Electrohysterogram