Articles in PresS. J Appl Physiol (December 30, 2014). doi:10.1152/japplphysiol.00144.2014
Very Long Apnea Events in Preterm Infants
Mary A. Mohr1*, Brooke D. Vergales2, Hoshik Lee1,3 Matthew T. Clark4†, Douglas E. Lake4,5, , Anne C. Mennen1, John Kattwinkel2, Robert A. Sinkin2, J. Randall Moorman4,6,7, Karen D. Fairchild2, John B. Delos1 Affiliations: 1
Department of Physics, College of William and Mary, Williamsburg, Virginia 23187-8795
2
Department of Pediatrics (Neonatology), University of Virginia, Charlottesville, Virginia
3
Samsung Advanced Institute of Technology, South Korea
4
Division of Cardiovascular Medicine, University of Virginia, Charlottesville, Virginia
5
Department of Statistics, University of Virginia, Charlottesville, Virginia
6
Department of Biomedical Engineering, University of Virginia, Charlottesville, Virginia
7
Department of Molecular Physiology, University of Virginia, Charlottesville, Virginia
*To whom correspondence should be addressed. †Current address: Adult Medical Predictive Devices, Diagnostics, and Displays LLC, Charlottesville, VA 22903. Running Head:
Very Long Apnea Events in Neonates
Corresponding Author:
Keywords:
Mary Mohr Physics Department, William and Mary Williamsburg, Virginia, 23187-8795
[email protected] 757 221 3511
apnea neonatal hypoxia bradycardia
1 Copyright © 2014 by the American Physiological Society.
ABSTRACT Apnea is nearly universal among very low birth weight (VLBW) infants, and the associated bradycardia and desaturation may have detrimental consequences. We describe here very long (>60 second) central apnea events (VLAs) with bradycardia and desaturation, discovered using a computerized detection system applied to our database of over 100 infant-years of electronic signals. Results: 86 VLAs occurred in 29 out of 335 VLBW infants. 18 of the 29 infants had a clinical event or condition possibly related to the VLA. Most VLAs occurred while infants were on nasal continuous positive airway pressure, supplemental oxygen and caffeine. Apnea alarms on the bedside monitor activated in 66% of events, on average 28 seconds after cessation of breathing. Bradycardia alarms activated late, on average 64 seconds after cessation of breathing. Prior to VLAs oxygen saturation was unusually high, and during VLAs oxygen saturation and heart rate fell unusually slowly. We give measures of the relative severity of VLAs, and theoretical calculations that describe the rate of decrease of oxygen saturation. Conclusions: Clinical – Very long apnea events with bradycardia and desaturation are not rare. Apnea alarms failed to activate for about 1/3 of VLAs. It appears that NICU personnel respond quickly to bradycardia alarms, but not consistently to apnea alarms. We speculate that more reliable apnea detection systems would improve patient safety in the NICU. Physiological – The slow decrease of oxygen saturation is consistent with a physiological model based on assumed high values of initial oxygen saturation.
2
INTRODUCTION More than two decades ago, Southall, et al. reported prolonged apnea and bradycardia events in preterm infants cared for in neonatal intensive care units (NICUs) (60). In their study of 14 babies observed for 24 hour periods, the authors were disturbed to report cessations of breathing as long as 213 seconds, and to report that extended apnea events were often not detected or reported. Present practices(13, 18, 36) hold that any cessation of breathing longer than 10 seconds, if accompanied by bradycardia and oxygen desaturation, is an event that should be considered clinically significant (as should any apnea longer than 20 seconds regardless of bradycardia or desaturation). Most events last less than half a minute, either self-resolving or terminating when a caregiver stimulates the infant in response to a monitor alarm. Very long apnea events may reflect abnormal physiology, impending illness, alarm failure, and/or failure of bedside clinicians to rescue infants from prolonged apneic spells. Nearly all very low birthweight infants (VLBW, 60 seconds accompanied by bradycardia (heart rate below 100 beats per minute) and oxygen desaturation (SpO280
100%
60−80
80%
10−20
SpO2 30−40
60% −20
60−80
>80
0 20 40 60 80 100 time from start of apnea (seconds)
120
a) 170bpm Heart Rate 100bpm
SpO2
100%
22%−min
80% 1
0
85 beats
Probabilty of Apnea
−100
−50
b)
0 50 time (seconds) Deficit vs. Apnea Duration
100
150
Deficit: HR(beats) O2 (%−minutes)
100 HR
80 60 40
O2
20 0 0
20
40 60 80 Duration of Apnea (seconds)
100
120
a) 1.6 1.4
frequency
1.2
VLAs
1 0.8
10−20
0.6
10−20
0.4 0.2 0
0.4
0.6
0.8
1
1.2 1.4 1.6 1.8 breath length (seconds)
2
2.2
2.4
median SpO2 (%)
b) 100
95 60−80 >80 90 −5
−4
−3 −2 −1 time to start of apnea (minutes)
0
1
100% 80%
SpO2 CI probability of apnea
1 0
0
10 20 30 40 50 time from start of apnea (seconds)
60
Table 1:
Gestational age (weeks) Birth weight (grams) ELBW Male Mechanical ventilation at any time Death before NICU discharge Length of NICU stay (days)
VLA (n=29)
No VLA (n=306)
P=
26.5 ± 1.9
27.5 ± 3.1
0.09
952 ± 249
988 ± 299
0.53
19 (66%) 14 (48%)
150 (49%) 154 (50%)
0.12 0.85
23 (79%)
214 (70%)
0.39
1 (3%)
31 (10%)
0.34
78 ± 27
63 ± 42
0.02
Table 2: # of infants
Clinical association
GA
#VLAs
PMA
1
Died 48h after last VLA
27
19
28-32
1
Post-hemorrhagic hydrocephalus, ventricular tap performed prior to VLA cluster
26
3
31
1
Grade IV intraventricular hemorrhage
26
2
30
1
On prostaglandins for congenital heart disease
29
3
31
2
On medication for severe GE reflux
25
3-12
27-29
23-29
1-19
26-31
25-30
1-19
25-32
23-29
1-4
24-31
24-31
1-5
27-34
4 10 7 11
Antibiotics started for suspected sepsis within 24h of VLA Receiving antibiotics at time of event Intubated within 12h of VLA None identified, other than prematurity
Table 3: Groups
Group I: Infants with Indisputable 60 second VLAs Group N: Infants who Never had a ABD 60 Infants who never had a computerdetected ABD10 (included in Group N) All VLBW Infants:
number of ABD10 tags
number of infants
% of all VLBW babies
Tags per babyday of data
6,617
29
9
4.29
6,300
199
59
1.52
0
62
19
0
30,793
335
100
2.94
Table 4: Initial Conditions and Constants values in parenthesis were obtained using the one-at-a-time method parameter
ABD 10-20
ABD 30-40
ABD 60-80 (VLAs)
ABD 80 (VLAs)
S
0.870 (0.005)
0.915 (0.007)
0.950 (0.005)
0.970 (0.003)
Sv
0.70 (0.01)
0.74 (0.01)
0.800 (0.005)
0.825 (0.005)
C
13.3 (0.6)
d
0.2 (0.6)
Tc
12 (1)