- Email: [email protected]
A definition of human fetal apnea and the distribution of fetal apneic intervals during the last ten weeks of pregnancy
A definition of human fetal apnea and the distribution of fetal apneic intervals during the last ten weeks of pregnancy JOHN
PATRICK,
M.D.*
KAREN
CAMPBELL,
M.E.Sc.
LESLEY
CARMICHAEL,
RENATO
NATALE,
BRYAN
RICHARDSON,
London,
Ontario,
B.Sc. M.D.** M.D.**
Canada
Fetal breathing movements were studied over 24-hour observation periods for a total of 744 hours in 31 healthy pregnant women at 30 to 39 weeks’ gestational age. An analysis of 165,766 breath-to-breath intervals in eight fetuses demonstrated that 97% of breath-to-breath intervals were less than 6 seconds in duration. It was concluded that a breath-to-breath interval of 6 seconds or more is a rational definition of fetal apnea during the last 10 weeks of pregnancy. An analysis of intervals during which time no fetal breathing activity occurred in the 31 fetuses demonstrated that prolonged episodes of apnea were distributed in a similar fashion at different gestational ages. The longest periods of apnea were 65 minutes at 30 to 31 weeks, 105 minutes at 34 to 35 weeks, and 120 minutes at 36 to 39 weeks. No periods of apnea of more than 45 minutes were measured during the second and third hours followirlg maternal meals. A useful strategy for recognition of significant fetal apnea would be to make observations during the second and third hours following maternal meals when apneic lengths greater than 45 minutes were not observed in this highly selected group of normal fetuses. (AM. J. OBSTET. GYNECOL. 136:471, 1960.)
MADE THE POINT THAT, in fetallambs,the presence of fetal breathing movements in utero was “a good indication of health, and more sensitive than changes in heart rate as a sign of impending trouble from a variety of pathologic causes.” Hypoxemia in fetal lambs diminished fetal breathing activity, and asphyxia was associated with the onset of fetal gasping movements.*. 3 Therefore, it seemed important to deDAWES'
of Obstetrics and Gynecology and of Western Ontario. by grants from the Canadian Medical Research
From the Departments Physiolog?, University
Supported Council and the Conn Smythe Research Crippled Chifdren.
Foundation
Sponsored
Investigation.
by the Society for Gynecologic
Reprint requests: Dr. John Patrick, Department of Obstetrics and Gynecology, St. Joseph’s Hospital, 268 Grosvenor St., London, Ontario, Canada N6A 4V2. *Canadian **Canadian
Medical Medical
0002-9378/80/040471+07$00.70/0@l
Research
Council
Research
Council
1980TheC.
Scholar. Fellows.
V.MosbyCo
for
termine whether fetal breathing activity might be useful as a measurement of human fetal health. Several groups have begun to examine the usefulness of fetal breathing movements during human pregnancy in separating healthy from sick fetuses in utero.4-S Because human fetal breathing movements are episodic, serious problems have arisen in interpretation of the significance of these preliminary data. It has been demonstrated that normal human fetuses may be apneic for periods of up to 122 minutes99 lo Therefore, clinical studies which have considered fetal apnea for periods of 10 to 30 minutes as abnormal may have included a significant group of normal fetuses in their reports. The purpose of this study was to determine a suitable definition for fetal apnea and to describe the normal variations in apneic lengths at different gestational ages during the last 10 weeks of pregnancy in healthy human fetuses. The underlying objective of this work was to develop strategies for design of clinical experi471
472
Patrick
February Am. J. Obstet.
et al
-192
HOURS
-165,786 -30-31 o - -
038-39
I
I
0
1
BREATHS WEEKS WEEKS
I
2
3 INTERVAL
I
5
4 LENGTH
(4pts) (4pts)
(SECS)
Fig. 1. Fetal breath-to-breath intervals were measured, and the histogram of the mean percentage of total breath-to-breath intervals in 0.5~second increments was plotted for four fetuses at 30 to 31 weeks (--o---) and four fetuses at 38 to 39 weeks (---a---).
28
WEEKS
r
24
20
16
12
8
4
I 0
I 20
I 40 INTERVAL
Fig. 2. Percentage weeks
was plotted
of intervals in 5-minute
with no fetal breathing intervals and multiples
I
I
80
60 LENGTH
I
I
100
120
(MIN)
movements (?SEM) in nine of 5-minute intervals.
fetuses
at 30 to 31
15. 1980 Gynecol.
Volume Number
1%
Fetal apnea
and distribution
of apneic
intervals
473
4
34-35
I cl
I 20
I
I
I
40
60
80
INTERVAL
Fig. 3. Percentage weeks
was plotted
of intervals in S-minute
ments in evaluation of the usefulness breathing movements.
with no fetal breathing intervals and multiples
of human
fetal
Methods Patients. Informed consent was obtained from nine healthy pregnant women at 30 to 3 1 weeks’ gestational age, 11 at 34 to 35 weeks, and 11 at 38 to 39 weeks. Fetal outcome confirmed gestational ages and the good health of fetuses studied. The complete patient data and the fetal outcome have been published elsewhere.s* lo Experimental design. The 31 patients were studied continuously for 24-hour observation periods for a total of 744 hours. Details of the experimental design have been published elsewhere.‘O Briefly, the patients rested in bed in a quiet room and were given 800 kcal meals at 0800, 1200, and 1700 hours. Neither patients nor observers were permitted to smoke during the study. Patients were studied continuously from 0800 to 0800 hours the following day, or from 2040 to 2000 hours the following day. Observations of fetuses were made with an ADR real-time ultrasonic scanner.* The *Model 2130, Tempe, Arizona.
Advanced
Diagnostic
Research
WEEKS
Corp.,
LENGTH
I 100
1 120
(MIN)
movements (&SEM) in 11 fetuses of S-minutes intervals.
at 34 to 35
transducer operated at a frequency of 3.5 MHz with an average intensity of 0.045 mW/sq cm. Individual fetal breathing movements were identified on a video monitor and recorded on a chart recorder using an event marke’r. The time during which gross fetal body movements were observed was indicated on the chart recorder using a second event marker. Analysis of fetal breathing movements and gross fetal body movements was performed on-line by a PDP 1 l/20* computer utilizing a program which was developed and validated for real-time data acquisition. The program also quantified fetal breathing movements, gross fetal body movements, and the time of patient interruption to the nearest 1 msec in real time. Measurements of the time during which fetal breathing movements occurred did not include episodes of apnea or gross fetal body movements. Gross fetal body movements were defined as fetal activity sufficient to prevent recognition of fetal breathing movements. Portions of records which could not be analyzed due to technical failure or patient interruptions were not included in the analysis and made up 6.5% t 0.8% of total time in the 30- to 3 l-week group, 8.7% ? 1.2% in *Digital Equipment
Corp., Maynard, Massachusetts.
474
Patrick et al.
February Am. J. Obstet.
38-39
INTERVAL
LENGTH
15, 1980 Gynecol.
WEEKS
(MIN)
Fig. 4. Percentage of intervals with no fetal breathing movements (tSEM) in I1 fetuses at 38 to 39 weeks was plotted in 5-minute intervals and multiples of S-minute intervals. the 34- to 35-week group, and 6.4% & O.j% of the time in the 38- to 39-week group. A validation of variations between observers in detcction of human fetal breathing movements has been published. I” The average percentage of variability among observers was 2.5%. Significance of data presented was determined by the use of a t test for small sample variance not assumed to be equal as described h\ Bailey.” Resutts ‘PO detcrniine a suitable interval ti)r definition of fetal apnea, 165.786 breath-to-breath intervals not interrupted by gross fetal body movements, technical failures, or patient interruptions were measured in 0%second intervals in a group of eight fetuses. Fig. I demonstra!es a histogram plot of the mean percentage of total breath-to-breath intervals in successive 0..5second increments in four fetuses at 30 to 31 weeks and in four fetuses at 38 to 39 weeks. At 30 to 31 weeks. there was a skewed distribution of breath-to-breath intervals with the peak to 0.5 to 1.O seconds. At 38 to 39
weeks, a similar skewed distribution was measured with a peak in breath-to-breath intervals at 1 .O to 1.5 seconds. This peak was considerably sharper than that for the 30- to Jl-week group. The mean rate of fetal breathing movements in the 30- to 31-week group was 58.3 2 3.7 breaths per minute, and this was significantly greater than the mean rate of 37 t 1.5 breaths per minute at 38 to 39 weeks (P < 0.05). At both 30 to 31 weeks and 38 to 39 weeks, there were onl) 3% of breath-to-breath intervals equal to or greater than 6 seconds. Therefore. at 30 to 31 weeks and 38 to 39 weeks a definition of fetal apnea of 6 seconds or more would only exclude 3$? of breath-to-breath intervals in this analysis of 165.786 breath-to-breath intervals. To determine the significance of prolonged apneic intervals, an analysis was performed on 288 successive .?-minute intervals in each of the 31 patients. The number of intervals of Fi-minute length during which time no fetal breathing activity occurred was determined in each patient and expressed as a percentage of the total number of intervals of that length during the studv period in each patient. The same analysis was
Volume Number
136 4
Fetal
28-
**e-
38-39 34-35
-*-
30-31
apnea
and distribution
of apneic
intervals
475
WEEKS I
24,
zo-
16,
12-
8,
4-
0,
I
I
I
I
I
0
20
40
60
80
INTERVAL
LENGTH
I 100
I 120
(MIN)
Fig. 5. Composite plot of the percentage of intervals with no fetal breathing movements in nine fetuses at 30 to. 31 weeks, 11 at 34 to 35 weeks, and 11 at 38 to 39 weeks demonstrating a similar distribution at the different gestational ages. The longest apneic interval was 65 minutes at 30 to 3 1 weeks, 105 minutes at 34 to 35 weeks, and 120 minutes at 38 to 39 weeks. carried out on intervals with lengths that were multiples of 5 minutes (10, 15, and 20 minutes, etc.). A plot of the percentage of 5-minute intervals or multiples of S-minute intervals during which time no fetal breathing movements occurred in nine patients at 30 to 3 1 weeks is demonstrated in Fig. 2. Twenty-seven percent of j-minute intervals contained no fetal breathing activity, and 21% of the lo-minute intervals contained no fetal breathing activity. The percentage of intervals containing no fetal breathing activity diminished steadily until only 3% of the intervals were greater than 45 minutes, and no apneic intervals were greater than 65 minutes. A similar plot of apneic intervals in the 11 fetuses at 34 to 35 weeks is seen in Fig. 3. Twenty-five percent of the 5-minute intervals contained no fetal breathing movements, less than 3% of the apneic intervals were greater than 55 minutes, and no apneic intervals were greater than 105 minutes. Similarly, a plot of apneic intervals in the 11 fetuses at 38 to 39 weeks is se& in Fig. 4. Twenty-six percent of the 5-minute intervals contained no fetal breathing
movements, 3% of the apneic intervals were more than 65 minutes, and no apneic intervals were greater than 120 minutes. A composite plot of Figs. 2 to 4 is shown in Fig. 5 which demonstrates that, at the three gestational ages, distribution of apneic intervals was very similar. The major difference was the length of the longest apneic interval which was 65 minutes at 30 to 31 weeks, 105 minutes at 34 to 35 weeks, and 120 minutes at 38 to 39 weeks. It has been demonstrated that at 30 to 3 1 weeks and 34 to 35 weeks there is a significant increase in fetal breathing activity during the second and third hours how this after maternal meals. y* lo Fig. 6 demonstrates effect changes the distribution of apneic intervals. This represents intervals from 11 fetuses at 34 to 35 weeks taken between 0900 to 1100, 1300 to 1500, and 1800 to 2000 hours which represented the second and third hours following meals. A similar plot of intervals in the same 11 fetuses between 0700 to 0900, 1100 to 1300, and 1600 to 1809 hours was made for comparison. During the second and third hours after meals, there
476
Patrick
February Am. J. Obstet.
et al.
- ,: -
15. 1980 C.ynecol.
BfFORE, , MEALS
Fig. 6. The percentage of intervals (tSEM) with no fetal breathing movements was significantly different during the second and third hours following meals (--•-) when compared to the hour I)ef’orc and during meals C---C+--) in 11 fetuses at 34 to 35 weeks (p < 0.01). were 1i.,S(;i- -L 2.02, of S-minute intervals with no fetal breathing activity. This percentage was significantly less than thar during the hour before and during meals when ther-e was 36% t 5.5% of 5-minute intervals with no fetal breathing activity (P < 0.01). TheI-e were no apneic intervals greater than 35 minutes during the srcond and third hours after meals which compared with 65 minutes during the 2 hours before and during meals. At 30 to 3 1 weeks, the longest period of apnea was 45 minutes during the second and third haul-s following meals as compared to 65 minutes during the hour befort and during meals. At 38 to 39 weeks, the longest period of apnea was 45 minutes during the second and third haul-s following meals as compared to 95 minutes during the hour before and during meals. Therefore, during the last 10 weeks of pregnancy. the longest period of time over which apnea occurred during the second and third hours following meals was 4j minutes.
Comment investigators have defined when 6 or more seconds
fetal apnea as a time when separate successive fetal
breathing movements.“. ‘. I” The present stud) confirmed that this arbitrary choice was reasonable. .4nalysis of 165,786 breath-to-breath intervals in four fetuses at 30 to 31 weeks and four at 38 to 39 weeks demonstrated that only 3% of breath-to-breath intervals were greater than or equal to 6 seconds. The distribution of breath-to-breath intervals was different at 30 to 31 weeks than at 38 to 39 weeks. At 30 to 31 weeks, breath-to-breath intervals peaked at 0.5 to 1.0 seconds and had a broad distribution about the mean. .4t 38 to 39 weeks. there was a sharply defined peak in breath-to-breath intervals at 1.0 to 1.5 seconds which suggested thar the more mature fetuses made regular breathing movements at a slower rate. Despite the gestational age difference, the choice of an apneic threshold of 6 seconds or more would seem to be rational as 97% of breath-to-breath intervals fell within this definition in the current study. Human and lamb fetuses during the last third of pregnancy make breathing movements in episodes about 30% to 40% of the tirne.le3’ 5, 6, y. ‘“3 ** Experimental or spontaneous hypoxemia in fetal lambs results in prolonged episodes of fetal apnea.‘, 3 During human pregnancy, absence of fetal breathing movements may
Volume Number
136 4
be a sign of fetal ill health, but normal fetuses ordinarily spend prolonged periods of time in apnea when observed with real-time scannerss, lo Campbell and associatesi3 have demonstrated that recurrence of episodes of fetal breathing movements are not random over 24-hour periods. Prolonged episodes of apnea are characteristic of healthy human fetuses in utero but present a major problem to clinical investigators who wish to design studies to quantify human fetal breathing movements as an indicator of fetal health. The problem facing investigators is, on one hand, to define a test which can be performed over a short enough period of time to be clinically feasible for widespread use and, on the other hand, to determine whether a given length of fetal apnea represents normal variation or an abnormal result. The data presented in this study suggest that random observations of fetal apnea of at least 2 hours would need to be made before a fetus could be said to fall outside the range of the normal population. Investigators might wish to further shorten the length of observation time while recognizing that a certain small percentage of the population might normally demonstrate apneic intervals greater than those measured as a result of the shorter observation time. For example, if random observation periods of 30 minutes are selected, it can be seen from Fig. 5 that there is only an 8%
Fetal
apnea
and distribution
of apneic
intervals
477
chance that apnea would occur over the entire interval in the normal population presented in this paper. A more useful strategy might be to perform fetal breathing measurements during the second and third hours following maternal meals (or a glucose load). If this strategy was adopted, measurements of fetal apnea for periods greater than 45 minutes would be outside the range of normal for the population described in this study. It is important to recognize that the data presented are an analysis of a highly selected group of normal pregnancies. Recent observations by Manning and associates’* that cigarette smoking decreases fetal breathing movements, by Fox and associates’5 that maternal alcohol consumption decreases fetal breathing activity for up to 1 hour, and by Richardson and associatesi that active labor causes an arrest of episodic breathing activity in healthy fetuses further emphasize the fact that much more information must be obtained on factors which normally influence fetal breathing activity before research strategies can be suggested for clinical evaluation of the usefulness of these measurements in the assessment of fetal health. We would like to thank Drs. J. Challis and P. Harding for their interest in this work, and Mrs. T. Clarke, Miss M, Harding, Messrs. J. Lemieux and D. Miller, and Ms. T. Vandenberg for their excellent technical assistance.
REFERENCES 1.
2.
3.
4.
9.
Dawes, G. S.: Breathing before birth in animals and man, N. Engl. J. Med. 290:557, 1974. Boddy, K., Dawes, G. S., Fisher, R., Pinter, S., and Robinson, J. S.: Foetal respiratory movements, electrocortical and cardiovascular responses to hypoxemia and hypercapnia in sheep, J. Physiol. (Land.) 243:599, 1974. Patrick, J. E., Dalton, K. J., and Dawes, G. S.: Breathing patterns before death in fetal lambs, AM. J. OBSTET. GYNECOL. 125:73, 1976. Manning, F. A., and Platt, L. D.: Fetal breathing movements and the abnormal contraction stress test, AM. J. OBSTET. GYNECOL. 133:590, 1979. Manning, F. A.: Fetal breathing movements as a reflection of fetal status. Posterad. Med. 61:116. 1977. Fox, H. E., and HohlerPC. W.: Fetal evaluation by realtime,imaging, Clin Obstet Gynecol. 20:339, 1977. Marsal, K.: Fetal breathing movements. Characteristics and clinical significance, Obstet. Gynecol. 52:394, 1978. Trudinger, B. J., Lewis, P. J., Mange&J., and O’Connor, E.: Fetal breathing movements in high risk pregnancy, Br. J. Obstet. Gynaecol. 85:662, 1978. Patrick, J., Campbell, K., Carmichael, L., Natale, R., and Richardson, B.: Patterns of human fetal breathing movements at 30 to 31 and 38 to 39 weeks’ gestational age, Obstet. Gynecol. In press.
10. Patrick, J., Natale, R., and Richardson, B.: Patterns of human fetal breathing activity at 34 to 35 weeks’ gestational age, AM. J. OB&ET. GYNECOL. 132:507, 1978. 11. Bailev. N. T. 1.: Statistical Methods in Biologv, London, 1973: English“LJniversities Press, p. 50. -’ 12. Patrick, J., Fetherston, W., Vick, H., and Voegelin, R.: Human fetal breathing movements and gross fetal body movements at weeks 34 to 35 of gestation, AM. J. OBSTET. GYNECOL. 130:693, 1978. 13. Campbell, K., MacNeill, I., and Patrick, J.: Time series analysis of human fetal breathing movements at 30-39 weeks gestational age, J. Biomed. Eng. In press. 14. Manning, F., Wyn Pugh, E., and Boddy, K.: Effect of cigarette smoking on fetal- breathing movements in normal pregnancies, Br. Med. J. 1:552, 1975. 15. Fox, H. E., Steinbrecher, M., Pessel, D., Inglis, J., Medvid, L., and Angel, E.: Maternal ethanol ingestion and the occurrence of human fetal breathing movements, AM. J. OBSTET. GYNECOL. 132:354, 1978. 16. Richardson, B., Natale, R., and Patrick, J.: Human fetal breathing activity during electively induced labor at term, AM. J. OBSTET. GYNECOL. 133:247, 1979.
PATRICK,
M.D.*
KAREN
CAMPBELL,
M.E.Sc.
LESLEY
CARMICHAEL,
RENATO
NATALE,
BRYAN
RICHARDSON,
London,
Ontario,
B.Sc. M.D.** M.D.**
Canada
Fetal breathing movements were studied over 24-hour observation periods for a total of 744 hours in 31 healthy pregnant women at 30 to 39 weeks’ gestational age. An analysis of 165,766 breath-to-breath intervals in eight fetuses demonstrated that 97% of breath-to-breath intervals were less than 6 seconds in duration. It was concluded that a breath-to-breath interval of 6 seconds or more is a rational definition of fetal apnea during the last 10 weeks of pregnancy. An analysis of intervals during which time no fetal breathing activity occurred in the 31 fetuses demonstrated that prolonged episodes of apnea were distributed in a similar fashion at different gestational ages. The longest periods of apnea were 65 minutes at 30 to 31 weeks, 105 minutes at 34 to 35 weeks, and 120 minutes at 36 to 39 weeks. No periods of apnea of more than 45 minutes were measured during the second and third hours followirlg maternal meals. A useful strategy for recognition of significant fetal apnea would be to make observations during the second and third hours following maternal meals when apneic lengths greater than 45 minutes were not observed in this highly selected group of normal fetuses. (AM. J. OBSTET. GYNECOL. 136:471, 1960.)
MADE THE POINT THAT, in fetallambs,the presence of fetal breathing movements in utero was “a good indication of health, and more sensitive than changes in heart rate as a sign of impending trouble from a variety of pathologic causes.” Hypoxemia in fetal lambs diminished fetal breathing activity, and asphyxia was associated with the onset of fetal gasping movements.*. 3 Therefore, it seemed important to deDAWES'
of Obstetrics and Gynecology and of Western Ontario. by grants from the Canadian Medical Research
From the Departments Physiolog?, University
Supported Council and the Conn Smythe Research Crippled Chifdren.
Foundation
Sponsored
Investigation.
by the Society for Gynecologic
Reprint requests: Dr. John Patrick, Department of Obstetrics and Gynecology, St. Joseph’s Hospital, 268 Grosvenor St., London, Ontario, Canada N6A 4V2. *Canadian **Canadian
Medical Medical
0002-9378/80/040471+07$00.70/0@l
Research
Council
Research
Council
1980TheC.
Scholar. Fellows.
V.MosbyCo
for
termine whether fetal breathing activity might be useful as a measurement of human fetal health. Several groups have begun to examine the usefulness of fetal breathing movements during human pregnancy in separating healthy from sick fetuses in utero.4-S Because human fetal breathing movements are episodic, serious problems have arisen in interpretation of the significance of these preliminary data. It has been demonstrated that normal human fetuses may be apneic for periods of up to 122 minutes99 lo Therefore, clinical studies which have considered fetal apnea for periods of 10 to 30 minutes as abnormal may have included a significant group of normal fetuses in their reports. The purpose of this study was to determine a suitable definition for fetal apnea and to describe the normal variations in apneic lengths at different gestational ages during the last 10 weeks of pregnancy in healthy human fetuses. The underlying objective of this work was to develop strategies for design of clinical experi471
472
Patrick
February Am. J. Obstet.
et al
-192
HOURS
-165,786 -30-31 o - -
038-39
I
I
0
1
BREATHS WEEKS WEEKS
I
2
3 INTERVAL
I
5
4 LENGTH
(4pts) (4pts)
(SECS)
Fig. 1. Fetal breath-to-breath intervals were measured, and the histogram of the mean percentage of total breath-to-breath intervals in 0.5~second increments was plotted for four fetuses at 30 to 31 weeks (--o---) and four fetuses at 38 to 39 weeks (---a---).
28
WEEKS
r
24
20
16
12
8
4
I 0
I 20
I 40 INTERVAL
Fig. 2. Percentage weeks
was plotted
of intervals in 5-minute
with no fetal breathing intervals and multiples
I
I
80
60 LENGTH
I
I
100
120
(MIN)
movements (?SEM) in nine of 5-minute intervals.
fetuses
at 30 to 31
15. 1980 Gynecol.
Volume Number
1%
Fetal apnea
and distribution
of apneic
intervals
473
4
34-35
I cl
I 20
I
I
I
40
60
80
INTERVAL
Fig. 3. Percentage weeks
was plotted
of intervals in S-minute
ments in evaluation of the usefulness breathing movements.
with no fetal breathing intervals and multiples
of human
fetal
Methods Patients. Informed consent was obtained from nine healthy pregnant women at 30 to 3 1 weeks’ gestational age, 11 at 34 to 35 weeks, and 11 at 38 to 39 weeks. Fetal outcome confirmed gestational ages and the good health of fetuses studied. The complete patient data and the fetal outcome have been published elsewhere.s* lo Experimental design. The 31 patients were studied continuously for 24-hour observation periods for a total of 744 hours. Details of the experimental design have been published elsewhere.‘O Briefly, the patients rested in bed in a quiet room and were given 800 kcal meals at 0800, 1200, and 1700 hours. Neither patients nor observers were permitted to smoke during the study. Patients were studied continuously from 0800 to 0800 hours the following day, or from 2040 to 2000 hours the following day. Observations of fetuses were made with an ADR real-time ultrasonic scanner.* The *Model 2130, Tempe, Arizona.
Advanced
Diagnostic
Research
WEEKS
Corp.,
LENGTH
I 100
1 120
(MIN)
movements (&SEM) in 11 fetuses of S-minutes intervals.
at 34 to 35
transducer operated at a frequency of 3.5 MHz with an average intensity of 0.045 mW/sq cm. Individual fetal breathing movements were identified on a video monitor and recorded on a chart recorder using an event marke’r. The time during which gross fetal body movements were observed was indicated on the chart recorder using a second event marker. Analysis of fetal breathing movements and gross fetal body movements was performed on-line by a PDP 1 l/20* computer utilizing a program which was developed and validated for real-time data acquisition. The program also quantified fetal breathing movements, gross fetal body movements, and the time of patient interruption to the nearest 1 msec in real time. Measurements of the time during which fetal breathing movements occurred did not include episodes of apnea or gross fetal body movements. Gross fetal body movements were defined as fetal activity sufficient to prevent recognition of fetal breathing movements. Portions of records which could not be analyzed due to technical failure or patient interruptions were not included in the analysis and made up 6.5% t 0.8% of total time in the 30- to 3 l-week group, 8.7% ? 1.2% in *Digital Equipment
Corp., Maynard, Massachusetts.
474
Patrick et al.
February Am. J. Obstet.
38-39
INTERVAL
LENGTH
15, 1980 Gynecol.
WEEKS
(MIN)
Fig. 4. Percentage of intervals with no fetal breathing movements (tSEM) in I1 fetuses at 38 to 39 weeks was plotted in 5-minute intervals and multiples of S-minute intervals. the 34- to 35-week group, and 6.4% & O.j% of the time in the 38- to 39-week group. A validation of variations between observers in detcction of human fetal breathing movements has been published. I” The average percentage of variability among observers was 2.5%. Significance of data presented was determined by the use of a t test for small sample variance not assumed to be equal as described h\ Bailey.” Resutts ‘PO detcrniine a suitable interval ti)r definition of fetal apnea, 165.786 breath-to-breath intervals not interrupted by gross fetal body movements, technical failures, or patient interruptions were measured in 0%second intervals in a group of eight fetuses. Fig. I demonstra!es a histogram plot of the mean percentage of total breath-to-breath intervals in successive 0..5second increments in four fetuses at 30 to 31 weeks and in four fetuses at 38 to 39 weeks. At 30 to 31 weeks. there was a skewed distribution of breath-to-breath intervals with the peak to 0.5 to 1.O seconds. At 38 to 39
weeks, a similar skewed distribution was measured with a peak in breath-to-breath intervals at 1 .O to 1.5 seconds. This peak was considerably sharper than that for the 30- to Jl-week group. The mean rate of fetal breathing movements in the 30- to 31-week group was 58.3 2 3.7 breaths per minute, and this was significantly greater than the mean rate of 37 t 1.5 breaths per minute at 38 to 39 weeks (P < 0.05). At both 30 to 31 weeks and 38 to 39 weeks, there were onl) 3% of breath-to-breath intervals equal to or greater than 6 seconds. Therefore. at 30 to 31 weeks and 38 to 39 weeks a definition of fetal apnea of 6 seconds or more would only exclude 3$? of breath-to-breath intervals in this analysis of 165.786 breath-to-breath intervals. To determine the significance of prolonged apneic intervals, an analysis was performed on 288 successive .?-minute intervals in each of the 31 patients. The number of intervals of Fi-minute length during which time no fetal breathing activity occurred was determined in each patient and expressed as a percentage of the total number of intervals of that length during the studv period in each patient. The same analysis was
Volume Number
136 4
Fetal
28-
**e-
38-39 34-35
-*-
30-31
apnea
and distribution
of apneic
intervals
475
WEEKS I
24,
zo-
16,
12-
8,
4-
0,
I
I
I
I
I
0
20
40
60
80
INTERVAL
LENGTH
I 100
I 120
(MIN)
Fig. 5. Composite plot of the percentage of intervals with no fetal breathing movements in nine fetuses at 30 to. 31 weeks, 11 at 34 to 35 weeks, and 11 at 38 to 39 weeks demonstrating a similar distribution at the different gestational ages. The longest apneic interval was 65 minutes at 30 to 3 1 weeks, 105 minutes at 34 to 35 weeks, and 120 minutes at 38 to 39 weeks. carried out on intervals with lengths that were multiples of 5 minutes (10, 15, and 20 minutes, etc.). A plot of the percentage of 5-minute intervals or multiples of S-minute intervals during which time no fetal breathing movements occurred in nine patients at 30 to 3 1 weeks is demonstrated in Fig. 2. Twenty-seven percent of j-minute intervals contained no fetal breathing activity, and 21% of the lo-minute intervals contained no fetal breathing activity. The percentage of intervals containing no fetal breathing activity diminished steadily until only 3% of the intervals were greater than 45 minutes, and no apneic intervals were greater than 65 minutes. A similar plot of apneic intervals in the 11 fetuses at 34 to 35 weeks is seen in Fig. 3. Twenty-five percent of the 5-minute intervals contained no fetal breathing movements, less than 3% of the apneic intervals were greater than 55 minutes, and no apneic intervals were greater than 105 minutes. Similarly, a plot of apneic intervals in the 11 fetuses at 38 to 39 weeks is se& in Fig. 4. Twenty-six percent of the 5-minute intervals contained no fetal breathing
movements, 3% of the apneic intervals were more than 65 minutes, and no apneic intervals were greater than 120 minutes. A composite plot of Figs. 2 to 4 is shown in Fig. 5 which demonstrates that, at the three gestational ages, distribution of apneic intervals was very similar. The major difference was the length of the longest apneic interval which was 65 minutes at 30 to 31 weeks, 105 minutes at 34 to 35 weeks, and 120 minutes at 38 to 39 weeks. It has been demonstrated that at 30 to 3 1 weeks and 34 to 35 weeks there is a significant increase in fetal breathing activity during the second and third hours how this after maternal meals. y* lo Fig. 6 demonstrates effect changes the distribution of apneic intervals. This represents intervals from 11 fetuses at 34 to 35 weeks taken between 0900 to 1100, 1300 to 1500, and 1800 to 2000 hours which represented the second and third hours following meals. A similar plot of intervals in the same 11 fetuses between 0700 to 0900, 1100 to 1300, and 1600 to 1809 hours was made for comparison. During the second and third hours after meals, there
476
Patrick
February Am. J. Obstet.
et al.
- ,: -
15. 1980 C.ynecol.
BfFORE, , MEALS
Fig. 6. The percentage of intervals (tSEM) with no fetal breathing movements was significantly different during the second and third hours following meals (--•-) when compared to the hour I)ef’orc and during meals C---C+--) in 11 fetuses at 34 to 35 weeks (p < 0.01). were 1i.,S(;i- -L 2.02, of S-minute intervals with no fetal breathing activity. This percentage was significantly less than thar during the hour before and during meals when ther-e was 36% t 5.5% of 5-minute intervals with no fetal breathing activity (P < 0.01). TheI-e were no apneic intervals greater than 35 minutes during the srcond and third hours after meals which compared with 65 minutes during the 2 hours before and during meals. At 30 to 3 1 weeks, the longest period of apnea was 45 minutes during the second and third haul-s following meals as compared to 65 minutes during the hour befort and during meals. At 38 to 39 weeks, the longest period of apnea was 45 minutes during the second and third haul-s following meals as compared to 95 minutes during the hour before and during meals. Therefore, during the last 10 weeks of pregnancy. the longest period of time over which apnea occurred during the second and third hours following meals was 4j minutes.
Comment investigators have defined when 6 or more seconds
fetal apnea as a time when separate successive fetal
breathing movements.“. ‘. I” The present stud) confirmed that this arbitrary choice was reasonable. .4nalysis of 165,786 breath-to-breath intervals in four fetuses at 30 to 31 weeks and four at 38 to 39 weeks demonstrated that only 3% of breath-to-breath intervals were greater than or equal to 6 seconds. The distribution of breath-to-breath intervals was different at 30 to 31 weeks than at 38 to 39 weeks. At 30 to 31 weeks, breath-to-breath intervals peaked at 0.5 to 1.0 seconds and had a broad distribution about the mean. .4t 38 to 39 weeks. there was a sharply defined peak in breath-to-breath intervals at 1.0 to 1.5 seconds which suggested thar the more mature fetuses made regular breathing movements at a slower rate. Despite the gestational age difference, the choice of an apneic threshold of 6 seconds or more would seem to be rational as 97% of breath-to-breath intervals fell within this definition in the current study. Human and lamb fetuses during the last third of pregnancy make breathing movements in episodes about 30% to 40% of the tirne.le3’ 5, 6, y. ‘“3 ** Experimental or spontaneous hypoxemia in fetal lambs results in prolonged episodes of fetal apnea.‘, 3 During human pregnancy, absence of fetal breathing movements may
Volume Number
136 4
be a sign of fetal ill health, but normal fetuses ordinarily spend prolonged periods of time in apnea when observed with real-time scannerss, lo Campbell and associatesi3 have demonstrated that recurrence of episodes of fetal breathing movements are not random over 24-hour periods. Prolonged episodes of apnea are characteristic of healthy human fetuses in utero but present a major problem to clinical investigators who wish to design studies to quantify human fetal breathing movements as an indicator of fetal health. The problem facing investigators is, on one hand, to define a test which can be performed over a short enough period of time to be clinically feasible for widespread use and, on the other hand, to determine whether a given length of fetal apnea represents normal variation or an abnormal result. The data presented in this study suggest that random observations of fetal apnea of at least 2 hours would need to be made before a fetus could be said to fall outside the range of the normal population. Investigators might wish to further shorten the length of observation time while recognizing that a certain small percentage of the population might normally demonstrate apneic intervals greater than those measured as a result of the shorter observation time. For example, if random observation periods of 30 minutes are selected, it can be seen from Fig. 5 that there is only an 8%
Fetal
apnea
and distribution
of apneic
intervals
477
chance that apnea would occur over the entire interval in the normal population presented in this paper. A more useful strategy might be to perform fetal breathing measurements during the second and third hours following maternal meals (or a glucose load). If this strategy was adopted, measurements of fetal apnea for periods greater than 45 minutes would be outside the range of normal for the population described in this study. It is important to recognize that the data presented are an analysis of a highly selected group of normal pregnancies. Recent observations by Manning and associates’* that cigarette smoking decreases fetal breathing movements, by Fox and associates’5 that maternal alcohol consumption decreases fetal breathing activity for up to 1 hour, and by Richardson and associatesi that active labor causes an arrest of episodic breathing activity in healthy fetuses further emphasize the fact that much more information must be obtained on factors which normally influence fetal breathing activity before research strategies can be suggested for clinical evaluation of the usefulness of these measurements in the assessment of fetal health. We would like to thank Drs. J. Challis and P. Harding for their interest in this work, and Mrs. T. Clarke, Miss M, Harding, Messrs. J. Lemieux and D. Miller, and Ms. T. Vandenberg for their excellent technical assistance.
REFERENCES 1.
2.
3.
4.
9.
Dawes, G. S.: Breathing before birth in animals and man, N. Engl. J. Med. 290:557, 1974. Boddy, K., Dawes, G. S., Fisher, R., Pinter, S., and Robinson, J. S.: Foetal respiratory movements, electrocortical and cardiovascular responses to hypoxemia and hypercapnia in sheep, J. Physiol. (Land.) 243:599, 1974. Patrick, J. E., Dalton, K. J., and Dawes, G. S.: Breathing patterns before death in fetal lambs, AM. J. OBSTET. GYNECOL. 125:73, 1976. Manning, F. A., and Platt, L. D.: Fetal breathing movements and the abnormal contraction stress test, AM. J. OBSTET. GYNECOL. 133:590, 1979. Manning, F. A.: Fetal breathing movements as a reflection of fetal status. Posterad. Med. 61:116. 1977. Fox, H. E., and HohlerPC. W.: Fetal evaluation by realtime,imaging, Clin Obstet Gynecol. 20:339, 1977. Marsal, K.: Fetal breathing movements. Characteristics and clinical significance, Obstet. Gynecol. 52:394, 1978. Trudinger, B. J., Lewis, P. J., Mange&J., and O’Connor, E.: Fetal breathing movements in high risk pregnancy, Br. J. Obstet. Gynaecol. 85:662, 1978. Patrick, J., Campbell, K., Carmichael, L., Natale, R., and Richardson, B.: Patterns of human fetal breathing movements at 30 to 31 and 38 to 39 weeks’ gestational age, Obstet. Gynecol. In press.
10. Patrick, J., Natale, R., and Richardson, B.: Patterns of human fetal breathing activity at 34 to 35 weeks’ gestational age, AM. J. OB&ET. GYNECOL. 132:507, 1978. 11. Bailev. N. T. 1.: Statistical Methods in Biologv, London, 1973: English“LJniversities Press, p. 50. -’ 12. Patrick, J., Fetherston, W., Vick, H., and Voegelin, R.: Human fetal breathing movements and gross fetal body movements at weeks 34 to 35 of gestation, AM. J. OBSTET. GYNECOL. 130:693, 1978. 13. Campbell, K., MacNeill, I., and Patrick, J.: Time series analysis of human fetal breathing movements at 30-39 weeks gestational age, J. Biomed. Eng. In press. 14. Manning, F., Wyn Pugh, E., and Boddy, K.: Effect of cigarette smoking on fetal- breathing movements in normal pregnancies, Br. Med. J. 1:552, 1975. 15. Fox, H. E., Steinbrecher, M., Pessel, D., Inglis, J., Medvid, L., and Angel, E.: Maternal ethanol ingestion and the occurrence of human fetal breathing movements, AM. J. OBSTET. GYNECOL. 132:354, 1978. 16. Richardson, B., Natale, R., and Patrick, J.: Human fetal breathing activity during electively induced labor at term, AM. J. OBSTET. GYNECOL. 133:247, 1979.