It is well established that small for gestational age (SGA) fetuses are at higher risk of perinatal morbidity and mortality compared to appropriate for gestational age (AGA) fetuses [1] [2] [3] . Consequently much work has investigated approaches to antenatal monitoring of SGA fetuses including assessment of electronic fetal heart rate (FHR) monitoring [4]. Adverse cardiovascular and neurological outcomes have been reported in SGA infants in the early neonatal period [2] [5].

There is growing evidence that children born SGA have impaired autonomic nervous system (ANS) function which contributes to impaired cardiovascular rhythmicity [6]. Clinical studies have provided strong evidence of an association between circadian rhythmicity and several disorders of the cardiovascular system [2].

From as early as 20–22 weeks to the end of pregnancy circadian changes have been described in AGA fetuses during various physiological activities. These include FHR and variability, FHR accelerations and fetal movements [7] [8] [9]. In addition, studies have also shown a direct relationship between maternal and fetal heart rates at various stages in pregnancy, suggesting that fetus is not autonomous in this respect [9] [10].

Few studies have examined whether SGA fetuses exhibit similar circadian variation in FHR patterns as AGA fetuses. In a previous report we employed a small, portable fetal electrocardiogram device (Monica AN24) to demonstrate circadian variation in FHR patterns in AGA fetuses whilst the women continued their-day-today activities [11]. We hypothesized that SGA fetus may exhibit altered circadian changes in FHR parameters compared to AGA fetuses. To evaluate our hypothesis, we analyzed 17 h ambulatory FHR data in a cohort of otherwise healthy pregnant women carrying SGA fetus in their home environment.

A prospective, non-randomized observational study approved by the North-West Preston NRES Research Ethics committee (15/NW/0278) was conducted at Jessop Wing Hospital in Sheffield, UK. The study group consisted of 31 women with singleton pregnancies between 24 and 40 weeks gestation and ultrasound-estimated fetal weight (EFW) less than 10th gestation centile [12]. Women recruited for the study were normotensive and suffered no pre-existing medical conditions. At the time of recording, women who smoked were advised to press an "event button" on the monitor, every time they smoked. For the purpose of analysis, all those 30 minutes epochs where the "event mark" for smoking occurred were excluded. None of the women drank alcohol or received no other medication at the time of evaluation. Patients with any fetal malformations, who were in active labor, and who had multiple pregnancies were excluded from the study.

After giving informed written consent, the fetal ECG monitor was fitted by placing five skin electrodes in standardized positions on the maternal abdomen [13]. When fully charged the device is capable of obtaining recordings lasting up to 20 hours [14]. Participants were allowed to go home and carry on with daily activities whilst wearing the monitor. They were advised to turn off and remove the monitor after 20 hours had elapsed. However, they were at liberty to take off the monitor at any time if they experienced any discomfort. The monitor was either collected from patient's home by the research team or returned by the patient after the monitoring session. Subsequently, the participants filled in a questionnaire to record their views and experience of wearing the device, and to express if they were willing to wear the device again.

The electrophysiological signal recorded within the monitor calculated beat-to-beat information on FHR in addition to the average FHR data obtained from Doppler-based CTG monitors. It also recorded the FHR parameters generated by computerized analysis developed by Dawes and Redman [15] [16].

In order to investigate changes in FHR pattern between the day and night times, three consecutive 30 minutes frames (90 minutes data) with FHR acquisition success rate of =80% were selected during the "day" and "night" periods for each subject. The three selected 30 min epochs were averaged and compared between the day and night recordings of the same individual.

The following FHR parameters derived by the Dawes Redman analysis were recorded and evaluated:

• Basal FHR, the baseline FHR, once the accelerations and decelerations are removed.
• Accelerations, defined as "a rise in FHR from the baseline greater than 10 bpm and lasting for more than 15 seconds".
• Decelerations, defined as "a fall in FHR from the baseline greater than 10 bpm and lasting for more than 15 seconds".
• Long-term variation (LTV), the mean of the difference between maximum and minimum FHR values in a one-minute sliding window
• Short-term variation (STV), the mean difference between adjacent 3.75 seconds FHR signals.

Statistical analysis
Fetal heart rate data were normally distributed and were described using the mean and standard deviation. In order to analyze the day and night recording of the same individual, the paired sample t-test was applied. The Pearson's rank-order correlation coefficient was used to study the relationship between FHR parameters and gestational age. An independent sample t-test was performed to determine if a difference in day and night FHR parameters was observed as a result of fetal gender, gestational age or smoking status.

Table 1 summarizes the basic demographic data for the women, study recording and delivery outcome variables.

62 recordings (31 days and 31 nights) made over 90 minute epochs were analyzed from 31 women recruited at home (Table 2). Significant differences were observed in the basal FHR and number of decelerations (p <0.001), both decreased at night time. However, STV, LTV and number of accelerations did not show significant variations between the day and night recordings (p>0.5).

Further analysis during the night-time, when external stimuli were reduced, showed the change in basal FHR to be strongly inversely correlated with gestational age; p=0.013 (Figure 1). However, other FHR parameters did not show a significant correlation with gestational age, (p>0.1). Since there was a clear effect of gestational age on basal FHR, the data were then subdivided into two groups. Table 3 demonstrates that fetuses below 34 weeks of gestation did not exhibit similar circadian changes in FHR patterns as mature fetuses did.

In our study, we had 10 women who smoked during the course of recording. Although, we excluded all the 30 minutes epochs where smoking occurred, we examined day: night changes in FHR parameters separately between smokers and non-smokers to assess whether maternal smoking status disrupts circadian changes in FHR pattern. Results given in Table 4 clearly indicates that compared to fetuses of non-smoking mothers, fetuses of smoking mothers did not show significant circadian variation in any of the FHR parameters. We did not observe any gender-related difference between 12 males and 19 females fetuses during the day and night time; p>0.1 Table 5.

Of the 31 women studied, 25 gave birth to babies with a birth weight below the tenth percentile.

This study has demonstrated that there is a clear circadian pattern in the FHR in SGA fetuses: there was a greater influence of time of day on basal FHR and number of decelerations, both of which reduced significantly at night time (p<0.001). However, unlike AGA fetuses [11], we did not observe significant circadian changes in FHR variation or number of acceleration in SGA fetuses. In addition, we did not find any difference in the day and night recordings between male and female fetuses as observed in AGA fetuses [11].

Maternal activity, diet intake, psychological and physiological conditions can influence FHR [17][18][19]. The fact that information regarding these maternal factors was not collected can be considered as a limitation of this study. However, the ability to achieve good quality recordings of the FHR over prolonged periods (up to 17 hours) in SGA fetuses whilst the mother continued with normal activities at home is a novel approach, permitting the assessment of the FHR under normal (uncontrolled) conditions. This is the first study to report changes seen in the FHR in relation to time of day in SGA pregnancies where maternal mobility was not restricted.

It is not possible to comment on whether the control of the circadian pattern in the fetus is influenced by maternal activity because we did not investigate the relationship of the FHR to that of the mother. However, the fact that advancing gestational age has an influence on the magnitude of the day:night difference in basal FHR supports the notion that there a major fetal component to control of the circadian rhythm of FHR in SGA fetuses.

Several studies have shown blunting circadian rhythm in pregnancies complicated by pre-eclampsia [20][21]. Disruption of circadian rhythmicity in women with pre-eclampsia may result in hampered development of fetal circadian rhythmicity. However, for the purpose of this study we excluded women with SGA fetuses who had pre-eclampsia. In addition, during evaluation, no woman received any medication that could have influenced FHR and FHR variability [22] [23] [24]. However, ten women smoked while they were monitored. Although, we excluded all 30 minutes epochs in the recordings where the "event marker" for smoking was pressed by the woman, our study was limited by the fact that we relied on women to provide information on smoking rather than making an objective assessment of tobacco exposure. Nicotine consumption has not only been shown to produce greater effect in women but also delays the diurnal peak of circadian heart rate pattern [25]. This may explain why fetuses of smoking mothers did not exhibit circadian changes for any of the FHR parameter.

Another limitation of the study was the inability to standardize the day recordings. This limitation was related to the success rate of the recording. The average overall success rate was 48.6%, which is lower than reported in previous studies [13] [16]. Graatsma et al. [13] reported 82% of their recordings had a signal quality of >60%. However, their study included fetuses of all sizes and was not limited to fetuses with estimated birth weights below the tenth percentile. Additionally, Graatsma et al. [13] standardized their recording time from 5 pm until 8 am, which meant that the majority of the recordings were made during resting hours, and therefore, fewer recordings were disturbed by maternal movements. In our study, the device was fitted after the clinic appointment, which meant that recording time started between 9 am and 5 pm. As a result, the amount of time at rest varied significantly between the recordings. A lower success rate meant that it was not possible to fully standardize the daytime recordings, therefore we randomly selected the best 90 min FHR data with good fetal signals at daytime, implying that our study may have suffered selection bias. Nonetheless, selection of 90 min FHR data guaranteed the analysis of high variation episodes as in the human fetus the cycle of active and quiet sleep ranges from 60–90 minutes [25]. However, a higher success rate at night (84.1%) meant that it was possible to fully standardize all recordings at night time.

Fetal heart rate patterns change markedly with gestation, with a decrease in basal FHR and an increase in FHR variability as pregnancy advances, reflecting maturation of the fetal autonomic nervous system and development of fetal behavioral pattern in AGA fetuses [26] [27]. Although, SGA fetuses exhibited reduction in basal FHR with advancing gestation and at nighttime, they did not show any change in FHR variation as observed in our previous work [11] and the work published by others on AGA fetuses [7] [9]. This finding can be interpreted as a reflection of immaturity of the cardiovascular autonomic control system in SGA fetuses and is consistent with the current knowledge that SGA children display significantly altered cardiovascular circadian rhythmicity [2] [6], and an impaired and an immature autonomic cardiovascular control, relating to an increased risk of cardiovascular disease [1] [5][27]. There is evidence that the development of cardiovascular disease is initiated by unfavorable conditions during intra-uterine life, implying that the cardiac autonomic balance may be permanently altered according to the concept of fetal programming [28].

The correlation of heart rate and variability with gender remains a matter of interest to many scientists. Previous studies that have examined the influence of gender on FHR variability in AGA fetuses have reported conflicting results. Lange et al. [29] suggested that gender does not need to be considered when assessing FHR variability, whereas Fleisher et al. [30] and Amorim-Costa et al. [31] reported significant gender-related differences in FHR parameters at various gestational ages. With regard to SGA pregnancies, there is insufficient data on the influence of gender on FHR pattern. Kwon et al. [4] found significant gender-related differences in the FHR variability in a study of 14 males and 15 females SGA fetuses during the last two-hour of labor preceding delivery. Our results differ from them, as we did not find any significant difference in FHR pattern between the SGA male and female fetuses. This discrepancy can be explained by experimental setting and sample size. We examined FHR during the antenatal period whilst Kwon et al. carried out studies during labor [4]. Secondly, we did not have equal distribution of male and female fetuses in our study cohort.

This study confirms that small for gestational age (SGA) fetuses demonstrate a clear circadian pattern in underlying basal fetal heart rate (FHR) and decelerations, both of which reduce significantly at nighttime. An analysis of these fluctuations is useful in the evaluation of the health of the fetus. In general clinical settings, FHR is usually monitored in the daytime. As the number of decelerations is higher in the daytime, we conclude that evaluation of fetal wellbeing is insufficient if done only during the day. Larger studies will be required to determine whether lack of circadian variation in FHR patterns in SGA fetuses correlates with adverse pregnancy indices and outcomes such as fetal acidosis. Nevertheless clinicians should take cognizance of these changes and exercise caution when applying and interpreting FHR tests for antenatal fetal wellbeing.

This study was funded by Jessop Wing Small Grant Scheme. We gratefully acknowledge Monica Healthcare, Nottingham, UK for providing electrodes in kind for the study.

AGA   Appropriate for gestational age
ANS   Autonomic nervous system
bpm    beats per minute
EFW    Expected fetal weight
FHR    Fetal heart rate
LTV    Long-term variation
ms     millisecond
SGA    Small for gestational age
STV    Short-term-variation

1. Galland BC, Taylor BJ, Bolton DP, Sayers RM. Heart rate variability and cardiac reflexes in small for gestational age infants. J Appl Physiol (1985) 2006 Mar;100(3):933–9.   [CrossRef]   [Pubmed]    
2. Begum EA, Bonno M, Sasaki N, et al. Blunted heart rate circadian rhythms in small for gestational age infants during the early neonatal period. Am J Perinatol 2012 May;29(5):369–76.   [CrossRef]   [Pubmed]    
3. Yanagihara T, Hata T. Comparison of late-second-trimester nonstress test characteristics between small for gestational age and appropriate for gestational age infants. Obstet Gynecol 1999 Dec;94(6):921–4.   [CrossRef]   [Pubmed]    
4. Kwon JY, Park IY, Lim J, Shin JC. Changes in spectral power of fetal heart rate variability in small-for-gestational-age fetuses are associated with fetal sex. Early Hum Dev 2014 Jan;90(1):9–13.   [CrossRef]   [Pubmed]    
5. Zamecznik A, Stanczyk J, Wosiak A, Niewiadomska-Jarosik K. Time domain parameters of heart rate variability in children born as small-for-gestational age. Cardiol Young 2016 Jul 26:1–8.   [CrossRef]   [Pubmed]    
6. Wolfenstetter A, Simonetti GD, Pöschl J, Schaefer F, Wühl E. Altered cardiovascular rhythmicity in children born small for gestational age. Hypertension 2012 Sep;60(3):865–70.   [CrossRef]   [Pubmed]    
7. de Vries JI, Visser GH, Mulder EJ, Prechtl HF. Diurnal and other variations in fetal movement and heart rate patterns at 20-22 weeks. Early Hum Dev 1987 Nov;15(6):333–48.   [CrossRef]   [Pubmed]    
8. Visser GH, Goodman JD, Levine DH, Dawes GS. Diurnal and other cyclic variations in human fetal heart rate near term. Am J Obstet Gynecol 1982 Mar 1;142(5):535–44.   [CrossRef]   [Pubmed]    
9. Lunshof S, Boer K, Wolf H, van Hoffen G, Bayram N, Mirmiran M. Fetal and maternal diurnal rhythms during the third trimester of normal pregnancy: outcomes of computerized analysis of continuous twenty-four-hour fetal heart rate recordings. Am J Obstet Gynecol 1998 Feb;178(2):247–54.   [CrossRef]   [Pubmed]    
10. Patrick J, Campbell K, Carmichael L, Natale R, Richardson B. Daily relationships between fetal and maternal heart rates at 38 to 40 weeks of pregnancy. Can Med Assoc J 1981 May 1;124(9):1177–8.   [Pubmed]    
11. Kapaya H, Broughton Pipkin F, Hayes-Gill B, Loughna PV. Circadian changes and sex-related differences in fetal heart rate parameters. Matern Health Neonatol Perinatol 2016 Sep 2;2(1):9.   [CrossRef]   [Pubmed]    
12. Gardosi J, Clausson B, Francis A. The value of customised centiles in assessing perinatal mortality risk associated with parity and maternal size. BJOG 2009 Sep;116(10):1356–63.   [CrossRef]   [Pubmed]    
13. Graatsma EM, Jacod BC, van Egmond LA, Mulder EJ, Visser GH. Fetal electrocardiography: Feasibility of long-term fetal heart rate recordings. BJOG 2009 Jan;116(2):334–7; discussion 337–8.   [CrossRef]   [Pubmed]    
14. Monica Healthcare, 2015. Monica AN24. [Available at: http://www.monicahealthcare.com/Monica_Healthcare/media/Monica/Products%20Sheets/Monica-AN24-IF24-Fetal-Monitor-Sales-Sheet.pdf]    
15. Kapaya H, Broughton-Pipkin F, Hayes-Gill B, Loughna PV. Smoking in pregnancy affects the fetal heart: Possible links to future cardiovascular disease. J Matern Fetal Neonatal Med 2015;28(14):1664–8.   [CrossRef]   [Pubmed]    
16. Stampalija T, Casati D, Montico M, et al. Parameters influence on acceleration and deceleration capacity based on trans-abdominal ECG in early fetal growth restriction at different gestational age epochs. Eur J Obstet Gynecol Reprod Biol 2015 May;188:104–12.   [CrossRef]   [Pubmed]    
17. Esin S. Factors that increase reactivity during fetal nonstress testing. Curr Opin Obstet Gynecol 2014 Apr;26(2):61–6.   [CrossRef]   [Pubmed]    
18. Muro M, Shono H, Shono M, Ito Y, Uchiyama A, Iwasaka T. Analysis of the influence of diurnal variation in maternal movements on fetal heart rate acceleration. Psychiatry Clin Neurosci 2002 Jun;56(3):247–8.   [CrossRef]   [Pubmed]    
19. Marzbanrad F, Kimura Y, Palaniswami M, Khandoker AH. Quantifying the interactions between maternal and fetal heart rates by transfer entropy. PLoS One 2015 Dec 23;10(12):e0145672.   [CrossRef]   [Pubmed]    
20. Koenen SV, Franx A, Mulder EJ, Bruinse HW, Visser GH. Fetal and maternal cardiovascular diurnal rhythms in pregnancies complicated by pre-eclampsia and intrauterine growth restriction. J Matern Fetal Neonatal Med 2002 May;11(5):313–20.   [CrossRef]   [Pubmed]    
21. Ditisheim AJ, Dibner C, Philippe J, Pechère-Bertschi A. Biological rhythms and preeclampsia. Front Endocrinol (Lausanne) 2013 Apr 8;4:47.   [CrossRef]   [Pubmed]    
22. Piazze J, Dillon KC, Albana C. Full-term-pregnancy effects of antenatal betamethasone administration on short-term variation as assessed by computerized cardiotocography. J Prenat Med 2012 Apr;6(2):18–21.   [Pubmed]    
23. Nensi A, De Silva DA, von Dadelszen P, et al. Effect of magnesium sulphate on fetal heart rate parameters: A systematic review. J Obstet Gynaecol Can 2014 Dec;36(12):1055–64.   [CrossRef]   [Pubmed]    
24. Adan A, Sánchez-Turet M. Influence of smoking and gender on diurnal variations of heart rate reactivity in humans. Neurosci Lett 2001 Jan 12;297(2):109–12.   [CrossRef]   [Pubmed]    
25. Muro M, Shono H, Kohara M, Ito Y, Uchiyama A, Sugimori H. Diurnal variations in resting-active cycles in full-term fetal heart rate changes. Early Hum Dev 1996 Jan 5;44(1):51–8.   [CrossRef]   [Pubmed]    
26. Brändle J, Preissl H, Draganova R, et al. Heart rate variability parameters and fetal movement complement fetal behavioral states detection via magnetography to monitor neurovegetative development. Front Hum Neurosci 2015 Apr 7;9:147.   [CrossRef]   [Pubmed]    
27. Park MI, Hwang JH, Cha KJ, Park YS, Koh SK. Computerized analysis of fetal heart rate parameters by gestational age. Int J Gynaecol Obstet 2001 Aug;74(2):157–64.   [CrossRef]   [Pubmed]    
28. Schäffer L, Burkhardt T, Müller-Vizentini D, et al. Cardiac autonomic balance in small-for-gestational-age neonates. Am J Physiol Heart Circ Physiol 2008 Feb;294(2):H884–90.   [CrossRef]   [Pubmed]    
29. Lange S, Van Leeuwen P, Geue D, Hatzmann W, Grönemeyer D. Influence of gestational age, heart rate, gender and time of day on fetal heart rate variability. Med Biol Eng Comput 2005 Jul;43(4):481–6.   [CrossRef]   [Pubmed]    
30. Fleisher LA, Dipietro JA, Johnson TR, Pincus S. Complementary and non-coincident increases in heart rate variability and irregularity during fetal development. Clin Sci (Lond) 1997 Apr;92(4):345–9.   [CrossRef]   [Pubmed]    
31. Amorim-Costa C, Cruz J, Ayres-de-Campos D, Bernardes J. Gender-specific reference charts for cardiotocographic parameters throughout normal pregnancy: A retrospective cross-sectional study of 9701 fetuses. Eur J Obstet Gynecol Reprod Biol 2016 Apr;199:102–7.   [CrossRef]   [Pubmed]