Original Article
839
Severe Intraventricular Hemorrhage in Extremely Premature Infants: Are high Carbon Dioxide Pressure or Fluctuations the Culprit? Dima Altaany, MD1
Girija Natarajan, MD1
Dhruv Gupta, MD2
1 Division of Neonatal Perinatal Medicine, Hutzel Women’s Hospital,
Wayne State University, Detroit, Michigan 2 Department of Pediatrics, Wayne State University, Detroit, Michigan
Marwan Zidan, PhD2
Sanjay Chawla, MD1
Address for correspondence Sanjay Chawla, MD, Division of Neonatology, Wayne State University, 3901 Beaubien, Detroit, MI 48201 (e-mail: [email protected]).
Abstract
Keywords
► intraventricular hemorrhage ► hypercapnia ► fluctuations in PCO2 ► carbon dioxide ► premature infants ► cerebral blood flow
Objective This study aims to examine the association between measures of hypercapnia and fluctuation in PCO2 and severe intraventricular hemorrhage (IVH) and to evaluate the prevalence of hypercapnia, hypocapnia, and fluctuations in PCO2 in the initial 72 hours of life among premature infants. Study Design Retrospective study of premature infants with birth weight < 1,250 g, who were receiving some respiratory support. All blood gases obtained in the first 3 days of life were collected. Univariate and multivariate analyses were performed to assess the association of hypercapnia, and fluctuations in PCO2 with severe IVH. Result Our cohort included 285 patients, of whom 84% were intubated. Only 20% patients had all blood gases in the normocapnia range; 9% had at least 1 gas with hypercapnia; 51% had at least 1 gas with hypocapnia, and 20% patients had both hypercapnia and hypocapnia at different times. Infants with severe IVH (n ¼ 41) had significantly higher peak PCO2 and greater fluctuations in PCO2 within a short interval, compared with those without severe IVH (n ¼ 227). After controlling for gestational age, gender, antenatal steroid exposure, presence of hypercapnia, and Apgar score at 5 minutes, fluctuation in PCO2 remained significantly associated with severe IVH. Conclusion Fluctuations in PCO2 within a short period may be more significantly associated with severe IVH than the mere presence of hypercapnia.
Severe intraventricular hemorrhage (IVH) in premature infants may be associated with poor neurodevelopmental outcomes.1–3 Animal and human studies have noted carbon dioxide pressure in the blood (PCO2) to be a regulator of cerebral blood flow (CBF). 4,5 Disturbed CBF may be an important pathologic factor in the development of IVH.6–8 Hypercapnia increases CBF with increased risk of IVH.9 Fluctuation in PCO2 also leads to fluctuation in the CBF,10 thereby predisposing to IVH.11 Most extremely premature infants require some assistance for breathing. While high PCO2 is mainly determined
by severity of lung disease, fluctuations in PCO2 may be due to multiple factors, including a change in respiratory compliance, development, or correction of endotracheal tube obstruction, lung atelectasis, or collapse, malposition of the endotracheal tube, improper initial ventilator settings, or adjustments of the respiratory support. Fluctuations in PCO2 and hypercapnia sometime may coexist in a premature infant on respiratory support. Our aim was to examine the association between measures of hypercapnia and fluctuation in PCO2 and severe IVH and to evaluate the prevalence of hypercapnia, hypocapnia, and
received June 30, 2014 accepted after revision November 14, 2014 published online January 21, 2015
Copyright © 2015 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1543950. ISSN 0735-1631.
Downloaded by: National University of Singapore. Copyrighted material.
Am J Perinatol 2015;32:839–844.
Intraventricular Hemorrhage in Premature
Altaany et al.
fluctuations in PCO2 in the initial 72 hours of life among extreme premature infants with birth weight < 1,250 g.
Methods Setting This was a retrospective cohort study of all liveborn infants at Hutzel Women’s Hospital, Detroit, MI, born between January 1, 2009 and December 31, 2011. The protocol was approved by the Institutional Review Board of Wayne State University.
Patients All liveborn infants with a birth weight of < 1,250 g needing respiratory support were included. Neonates who were not resuscitated at birth due to previability were excluded from the study. Respiratory support was defined as the need for continuous positive airway pressure (CPAP) or mechanical ventilation.
Data Collection Data collection included race, histological chorioamnionitis, antenatal steroid (ANS) administration, gestational age, birth weight, mode of delivery, gender, Apgar score at 5 minutes, need for delivery room resuscitation, and neonatal morbidity, including IVH. All blood gases done within the first 72 hours of postnatal life and the respiratory support were recorded.
Definitions We examined three measures of hypercapnia: (1) Hypercapnia was defined as any PCO2 65 mm Hg in the first 3 days of life. (2) Time weighted PCO2 (a reflection of average PCO2 in the first 3 days) was calculated using the formula described by Fabres et al.11 To calculate the time-weighted PCO2, we multiplied the PCO2 value with the time interval from birth for first gas/or from the previous blood gas in hours. These numbers were added together and divided by the postnatal age at last blood gas or 72 hours, whichever was earlier. (3) Cumulative hypercapnia index to hypercapnia was modified from previous definitions and with the consideration of both the duration and severity of hypercapnia. Whenever PCO2 was 65 mm Hg, the difference (PCO2 65) was multiplied by the duration of hypercapnia taken as the interval between the blood gas 65 and the preceding blood gas.12 All these values were added to calculate the cumulative index of hypercapnia exposure. This was divided by the age of patient in hours when last blood gas was obtained to yield the cumulative hypercapnia index per hour of life. Hypocapnia was defined as any PCO2 < 35 mm Hg. Normocapnia was defined as maintenance of all PCO2 values between 35 and 64 mm Hg in the first 3 days of life. Fluctuation in PCO2 was defined as the maximum difference between two consecutive PCO2 values within a 6-hour interval for each infant. We chose this short interval of 6 hours empirically to evaluate relatively quick fluctuations of PCO2 value in a patient. To evaluate the response of health care professionals to severe hypocapnia (blood gas with PCO2 < 30 mm Hg), we collected American Journal of Perinatology
Vol. 32
No. 9/2015
information on both the timing of attempted change on the ventilator, as well as time to reach PCO2 > 30 mm Hg. Patients were assigned the maximum form of respiratory support they received in the first 3 days of life. Occurrence of severe IVH was defined as grade III or IV IVH.13 In our center, preterm infants are resuscitated in the delivery room using blended oxygen, starting at 30% and adjusting based on the pulse oximeter readings. A T-piece resuscitator is available for delivery of CPAP or positive pressure ventilation. The initial dose of surfactant (Survanta [AbbVie Inc., North Chicago, IL]) is administered only for infants who are intubated and mechanically ventilated. Subsequent doses are administered based on the need for supplemental oxygen and high mean airway pressure on the ventilator. Ventilator adjustments are based on blood gases, which are typically performed every 4 to 6 hours. We use a permissive hypercarbia strategy, allowing PCO2 to be in the 45 to 64 mm Hg range as long as pH is > 7.20.
Statistical Analysis Descriptive statistical analyses were performed using the SPSS software version 19.0 (SPSS Inc., Chicago, IL). Results were expressed as mean ( standard deviation), median (range), or n (%) as appropriate. Bivariate comparisons of hypercapnia measures and fluctuations were performed using the t-test for continuous variables or Mann–Whitney U test as appropriate and chi-square test for categorical variables. Binary logistic regression analysis was performed to examine the associations between fluctuation in PCO2, and severe IVH while controlling for gestational age, ANS exposure, gender, Apgar score at 5 minutes, and hypercapnia. Higher gestational age and ANS exposure have previously been noted to be associated with the reduction in IVH.14 Apgar score at 5 minutes and the presence of hypercapnia (as a marker of respiratory severity) were also controlled for in the regression model. A p-value < 0.05 was considered significant.
Results Study Population During the study period, there were 342 live births with birth weight < 1,250 g. A total of 37 patients received only comfort care or died in the delivery room. Overall 20 patients received only nasal cannula or no respiratory support. A total of 285 neonates were included in the analysis for description of blood gas parameters. Noninvasive respiratory support, including CPAP or nasal intermittent positive pressure ventilation was the maximum form of respiratory support in 46 (16%) patients, while 239 (84%) were intubated and received mechanical ventilation. ►Table 1 describes the baseline characteristics and measures of hypercapnia, hypocapnia, and fluctuations in PCO2 within 72 hours in the study population. On initial blood gas, rates of hypocapnia, hypercapnia, and normocapnia were 14.4, 13, and 72.6%, respectively. Fluctuation in PCO2 10 was seen in 30.2%, 11 to 20 in 39.6%, and > 20 in 30.2%. The median time to wean the ventilator support following hypocapnia on the blood gas was 20 minutes. The median time for PCO2 to reach above 30 mm Hg was 4 hours.
Downloaded by: National University of Singapore. Copyrighted material.
840
Intraventricular Hemorrhage in Premature
Altaany et al.
841
Table 1 Baseline characteristics and ventilation profile
a
Gestational age (wk)
27 (25–28)
Birth weight (g)
890 (706–1,090)
Race (African American)
223 (82%)
Males
134 (47%)
Complete course of antenatal steroids
178 (62%)
Vaginal delivery
92 (32%)
Histologic chorioamnionitis
119 (46%)
Apgar score at 5 min < 5
50 (18%)
Delivery room intubation
183 (66%)
Time of first gas (h)
1 (1–4)
Number of gases in first 72 h
14 (11–16)
Exclusive hypocapnia (PCO2 < 35)
145 (51%)
Exclusive hypercapnia (PCO2 65)
25 (9%)
Normocapnia
57 (20%)
Any hypocapnia
195 (68%)
Any hypercapnia
75 (26%)
Both hypercapnia and hypocapnia
58 (20%)
Time weighted PCO2 mm Hg
40 (37–45)
Cumulative hypercapnia index in first 72 h of life
36 (0–99)
Cumulative hypercapnia index per h of life
0 (0–0.2)
Highest fluctuation in PCO2 mm Hg
14 (10–22)
Median (interquartile range) or n (%).
A total of 43 (15%) patients died of whom 17 did not have head ultrasound before death and were excluded from further analysis, leaving total of 268 patients in the analysis for severe IVH. Severe IVH (grade 3 or 4) was diagnosed in 41 (15%). Among patients, who had large fluctuation in PCO2, the rate of severe IVH was 23%, compared with 11% in patients, without large fluctuation in PCO2 (p ¼ 0.02). We further explored the association of severe IVH with known factors including gestation, gender, ANSs exposure, Apgar score at 5 minutes, and PCO2 derangements in a bivariate model (►Table 2). Low gestational age, lack of complete course of ANS, male gender, low Apgar score, highest PCO2, and maximum difference in PCO2 were significantly different in groups with and without severe IVH. A binary logistic regression analysis for severe IVH, controlling for gestational age, gender, complete course of ANS, Apgar score at 5 minutes (< 5), and hypercapnia (as a surrogate for severity of respiratory illness) noted that male gender, ANSs, and maximum difference in PCO2 were associated with severe IVH (►Table 3).
Discussion This single center study of a recent cohort of extreme premature infants requiring respiratory support provides some insight on the current clinical practices in PCO2 targeting.
The finding that only 20% of infants were in the target carbon dioxide range is consistent with one previous study noting poor compliance with the target oxygen saturations.15 The nurse–patient ratio, extent of respiratory support, and postmenstrual age were factors associated with achievement of target oxygen saturation range. We speculate that achievement of target carbon dioxide levels is potentially modifiable by some of these factors. We noted a high incidence of both hypercapnia and hypocapnia, which often coexisted. Hypocapnia was encountered in over 60% of the patients at one point in the first 3 days of life. Although, hypocapnia was not associated in this study or from previous reports with severe IVH, hypocapnia is known to be related to poor neurodevelopmental outcomes.12,16 This may be related to cerebral vasoconstriction in response to low PCO2 with subsequent cerebral ischemia.10 As hypocapnia is iatrogenic in most cases, it may be amenable to improvement with change in current practices. We examined various measures of hypercapnia in relation to severe IVH. Time weighted PCO2 was not noted to be associated with severe IVH. This supports earlier findings of Fabres et al. We speculate this due to a neutralizing effect of hypocapnia and hypercapnia with their frequent coexistence in the same subject (►Table 2). We then explored the relationship of the cumulative index of hypercapnia exposure and severe IVH. This again was also American Journal of Perinatology
Vol. 32
No. 9/2015
Downloaded by: National University of Singapore. Copyrighted material.
Variablea
Intraventricular Hemorrhage in Premature
Altaany et al.
Table 2 Factors associated with severe intracranial hemorrhage (grade III or IV) Severe ICH present n ¼ 41
Variable
a
Severe ICH absent n ¼ 227
p-Value
Gestational age (wk)
25.6 1.9
27 2.5
< 0.01
Birth weight (g)
862 223
899 220
0.34
Complete antenatal steroids course
13 (32%)
158 (70%)
< 0.01
Vaginal delivery
15 (37%)
73 (32%)
0.58
Male gender
26 (63%)
98 (43%)
0.02
Histologic chorioamnionitis
18 (56%)
98 (46%)
0.34
Apgar < 5 at 5 min
14 (37%)
31 (13.7%)
< 0.01
Highest PCO2
64.2 16.1
57.9 16
0.02
Any hypercapnia 65
14 (34%)
49 (21.5%)
0.11
Max difference in PCO2 (mm Hg)
24 17.1
16 12.6
< 0.01
Maximum difference > 10
35 (85%)
146 (65%)
0.07
Maximum difference > 20
20 (49%)
54 (24%)
0.02
Time weighted PCO2 mm Hg
40.7 4.8
40.7 6.3
0.88
Cumulative hypercapnia index mm Hg per h
0 (0–485)
0 (0–462)
0.08
Cumulative hypercapnia index per h of life
0 (0–6.8)
0 (0–6.4)
0.076
Mean (SD), median (range) or n (%).
Table 3 Logistic regression model for severe intracranial hemorrhage Variable
Odds ratio
95% confidence interval
p-Value
Gestation
0.83
0.69–1.01
0.05
Complete ANS
0.22
0.1–0.48
< 0.01
Male gender
2.5
1.2–5.4
0.015
5 minute Apgar score < 5
1.15
0.43–3.03
0.79
Hypercapnia
1.24
0.43–3.6
0.69
1.3
1.03–1.6
0.033
Max difference in PCO2 a
a
For every difference of 10 mm Hg between consecutive gases.
not significantly different in patients with and without severe IVH. Duration of hypercapnia may not be a significant factor associated with severe IVH. Hypercapnia and its association with severe IVH have been previously noted in few studies.9,17–20 Our study also confirmed this association. Fluctuations in PCO2 have also been noted to affect CBF with loss of cerebral autoregulation and increased risk of IVH11 and poor neurodevelopmental outcomes.21 In the current study, we explored the effect of fluctuation in PCO2 within a short interval on the risk of severe IVH. We chose a maximum of 6 hours interval between two consecutive blood gases to study effect of relatively acute fluctuations in PCO2. Our current neonatal intensive care unit protocol is to obtain blood gases every 4 to 6 hours in the first 3 days of life in extreme premature infants. To the best of our knowledge, this is the first study that describes the role of fluctuation in PCO2 within such short interval and its relation to severe IVH. Fabres et al in an earlier report have described the effect of PCO2 fluctuations and severe IVH. This was described using the magnitude of standard deviation around American Journal of Perinatology
Vol. 32
No. 9/2015
the mean PCO2 levels as well as the absolute difference between highest and lowest PCO2 levels at any point in the first 96 hour as markers of fluctuation. In our model, however, we examined the effect of PCO2 dispersion more acutely, by assessing fluctuation in PCO2 within 6 hours at most. After adjustment for gestational age, gender, ANS, and hypercapnia, fluctuation in PCO2 of 10 mm Hg in the first 3 days of life was associated with an increase in the risk of severe IVH (odds ratio 1.3). Interestingly, hypercapnia lost significance in this model. This may imply that fluctuations in PCO2 may be more significantly associated with severe IVH than the mere presence of hypercapnia. A major limitation in our study was the timing of blood gases. Frequency of obtaining blood gases was not standardized among all patients. The timing of the blood gases was determined by the clinical team, although our usual practice was to obtain blood gases every 4 to 6 hours in first 3 days of life. Sicker infants likely had more blood gases obtained, compared with neonates on lower respiratory support. Rates of hypercapnia and degree of fluctuation may be different
Downloaded by: National University of Singapore. Copyrighted material.
842
when PCO2 is monitored continuously. Arterial blood was obtained for analysis in 167 (58%) of the cohort. Arterialized capillary samples were used in absence of arterial lines. Due to this observational design of the study, there may be other unknown confounders associated with hypercapnia/fluctuations in PCO2 which may also predispose to severe IVH. A randomized controlled trial to study the effects of hypercapnia and rapid fluctuation in PCO2 in humans is not ethical.
7
8 9
Conclusion Normocapnia in the first 3 days of life in premature infants needing respiratory support was uncommon in our patient cohort with high rates of both hypercapnia and hypocapnia. Fluctuation in PCO2 was significantly associated with severe IVH. Continuous forms of PCO2 monitoring, such as transcutaneous PCO2 or end tidal CO2 monitoring may help reduce the incidence of these derangements and improve outcomes. Future studies using continuous forms of PCO2 monitoring may result in a better understanding of the role of hypercapnia and fluctuations in PCO2 in the development of severe IVH.
10
11
12
13
Conflict of Interest The authors declare no conflict of interest. 14
References
15
1 Futagi Y, Toribe Y, Ogawa K, Suzuki Y. Neurodevelopmental
2
3
4
5
6
outcome in children with intraventricular hemorrhage. Pediatr Neurol 2006;34(3):219–224 Bassan H, Feldman HA, Limperopoulos C, et al. Periventricular hemorrhagic infarction: risk factors and neonatal outcome. Pediatr Neurol 2006;35(2):85–92 Sherlock RL, Anderson PJ, Doyle LW; Victorian Infant Collaborative Study Group. Neurodevelopmental sequelae of intraventricular haemorrhage at 8 years of age in a regional cohort of ELBW/very preterm infants. Early Hum Dev 2005;81(11):909–916 Pryds O, Greisen G. Effect of PaCO2 and haemoglobin concentration on day to day variation of CBF in preterm neonates. Acta Paediatr Scand Suppl 1989;360:33–36 Hansen NB, Brubakk AM, Bratlid D, Oh W, Stonestreet BS. The effects of variations in PaCO2 on brain blood flow and cardiac output in the newborn piglet. Pediatr Res 1984;18(11): 1132–1136 Perlman JM, Goodman S, Kreusser KL, Volpe JJ. Reduction in intraventricular hemorrhage by elimination of fluctuating cere-
16
17
18
19 20 21
Altaany et al.
bral blood-flow velocity in preterm infants with respiratory distress syndrome. N Engl J Med 1985;312(21):1353–1357 Van Bel F, Van de Bor M, Stijnen T, Baan J, Ruys JH. Aetiological rôle of cerebral blood-flow alterations in development and extension of peri-intraventricular haemorrhage. Dev Med Child Neurol 1987;29(5):601–614 Ballabh P. Intraventricular hemorrhage in premature infants: mechanism of disease. Pediatr Res 2010;67(1):1–8 Kaiser JR, Gauss CH, Pont MM, Williams DK. Hypercapnia during the first 3 days of life is associated with severe intraventricular hemorrhage in very low birth weight infants. J Perinatol 2006; 26(5):279–285 Wyatt JS, Edwards AD, Cope M, et al. Response of cerebral blood volume to changes in arterial carbon dioxide tension in preterm and term infants. Pediatr Res 1991;29(6):553–557 Fabres J, Carlo WA, Phillips V, Howard G, Ambalavanan N. Both extremes of arterial carbon dioxide pressure and the magnitude of fluctuations in arterial carbon dioxide pressure are associated with severe intraventricular hemorrhage in preterm infants. Pediatrics 2007;119(2):299–305 Shankaran S, Langer JC, Kazzi SN, Laptook AR, Walsh M; National Institute of Child Health and Human Development Neonatal Research Network. Cumulative index of exposure to hypocarbia and hyperoxia as risk factors for periventricular leukomalacia in low birth weight infants. Pediatrics 2006;118(4):1654–1659 Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978; 92(4):529–534 Chawla S, Natarajan G, Rane S, Thomas R, Cortez J, Lua J. Outcomes of extremely low birth weight infants with varying doses and intervals of antenatal steroid exposure. J Perinat Med 2010;38(4):419–423 Sink DW, Hope SA, Hagadorn JI. Nurse:patient ratio and achievement of oxygen saturation goals in premature infants. Arch Dis Child Fetal Neonatal Ed 2011;96(2):F93–F98 Collins MP, Lorenz JM, Jetton JR, Paneth N. Hypocapnia and other ventilation-related risk factors for cerebral palsy in low birth weight infants. Pediatr Res 2001;50(6):712–719 Vela-Huerta MM, Amador-Licona M, Medina-Ovando N, AldanaValenzuela C. Factors associated with early severe intraventricular haemorrhage in very low birth weight infants. Neuropediatrics 2009;40(5):224–227 Thome UH, Ambalavanan N. Permissive hypercapnia to decrease lung injury in ventilated preterm neonates. Semin Fetal Neonatal Med 2009;14(1):21–27 Zhou W, Liu W. Hypercapnia and hypocapnia in neonates. World J Pediatr 2008;4(3):192–196 Fritz KI, Delivoria-Papadopoulos M. Mechanisms of injury to the newborn brain. Clin Perinatol 2006;33(3):573–591, v v McKee LA, Fabres J, Howard G, Peralta-Carcelen M, Carlo WA, Ambalavanan N. PaCO2 and neurodevelopment in extremely low birth weight infants. J Pediatr 2009;155(2):217–21.e1
American Journal of Perinatology
Vol. 32
No. 9/2015
843
Downloaded by: National University of Singapore. Copyrighted material.
Intraventricular Hemorrhage in Premature
839
Severe Intraventricular Hemorrhage in Extremely Premature Infants: Are high Carbon Dioxide Pressure or Fluctuations the Culprit? Dima Altaany, MD1
Girija Natarajan, MD1
Dhruv Gupta, MD2
1 Division of Neonatal Perinatal Medicine, Hutzel Women’s Hospital,
Wayne State University, Detroit, Michigan 2 Department of Pediatrics, Wayne State University, Detroit, Michigan
Marwan Zidan, PhD2
Sanjay Chawla, MD1
Address for correspondence Sanjay Chawla, MD, Division of Neonatology, Wayne State University, 3901 Beaubien, Detroit, MI 48201 (e-mail: [email protected]).
Abstract
Keywords
► intraventricular hemorrhage ► hypercapnia ► fluctuations in PCO2 ► carbon dioxide ► premature infants ► cerebral blood flow
Objective This study aims to examine the association between measures of hypercapnia and fluctuation in PCO2 and severe intraventricular hemorrhage (IVH) and to evaluate the prevalence of hypercapnia, hypocapnia, and fluctuations in PCO2 in the initial 72 hours of life among premature infants. Study Design Retrospective study of premature infants with birth weight < 1,250 g, who were receiving some respiratory support. All blood gases obtained in the first 3 days of life were collected. Univariate and multivariate analyses were performed to assess the association of hypercapnia, and fluctuations in PCO2 with severe IVH. Result Our cohort included 285 patients, of whom 84% were intubated. Only 20% patients had all blood gases in the normocapnia range; 9% had at least 1 gas with hypercapnia; 51% had at least 1 gas with hypocapnia, and 20% patients had both hypercapnia and hypocapnia at different times. Infants with severe IVH (n ¼ 41) had significantly higher peak PCO2 and greater fluctuations in PCO2 within a short interval, compared with those without severe IVH (n ¼ 227). After controlling for gestational age, gender, antenatal steroid exposure, presence of hypercapnia, and Apgar score at 5 minutes, fluctuation in PCO2 remained significantly associated with severe IVH. Conclusion Fluctuations in PCO2 within a short period may be more significantly associated with severe IVH than the mere presence of hypercapnia.
Severe intraventricular hemorrhage (IVH) in premature infants may be associated with poor neurodevelopmental outcomes.1–3 Animal and human studies have noted carbon dioxide pressure in the blood (PCO2) to be a regulator of cerebral blood flow (CBF). 4,5 Disturbed CBF may be an important pathologic factor in the development of IVH.6–8 Hypercapnia increases CBF with increased risk of IVH.9 Fluctuation in PCO2 also leads to fluctuation in the CBF,10 thereby predisposing to IVH.11 Most extremely premature infants require some assistance for breathing. While high PCO2 is mainly determined
by severity of lung disease, fluctuations in PCO2 may be due to multiple factors, including a change in respiratory compliance, development, or correction of endotracheal tube obstruction, lung atelectasis, or collapse, malposition of the endotracheal tube, improper initial ventilator settings, or adjustments of the respiratory support. Fluctuations in PCO2 and hypercapnia sometime may coexist in a premature infant on respiratory support. Our aim was to examine the association between measures of hypercapnia and fluctuation in PCO2 and severe IVH and to evaluate the prevalence of hypercapnia, hypocapnia, and
received June 30, 2014 accepted after revision November 14, 2014 published online January 21, 2015
Copyright © 2015 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1543950. ISSN 0735-1631.
Downloaded by: National University of Singapore. Copyrighted material.
Am J Perinatol 2015;32:839–844.
Intraventricular Hemorrhage in Premature
Altaany et al.
fluctuations in PCO2 in the initial 72 hours of life among extreme premature infants with birth weight < 1,250 g.
Methods Setting This was a retrospective cohort study of all liveborn infants at Hutzel Women’s Hospital, Detroit, MI, born between January 1, 2009 and December 31, 2011. The protocol was approved by the Institutional Review Board of Wayne State University.
Patients All liveborn infants with a birth weight of < 1,250 g needing respiratory support were included. Neonates who were not resuscitated at birth due to previability were excluded from the study. Respiratory support was defined as the need for continuous positive airway pressure (CPAP) or mechanical ventilation.
Data Collection Data collection included race, histological chorioamnionitis, antenatal steroid (ANS) administration, gestational age, birth weight, mode of delivery, gender, Apgar score at 5 minutes, need for delivery room resuscitation, and neonatal morbidity, including IVH. All blood gases done within the first 72 hours of postnatal life and the respiratory support were recorded.
Definitions We examined three measures of hypercapnia: (1) Hypercapnia was defined as any PCO2 65 mm Hg in the first 3 days of life. (2) Time weighted PCO2 (a reflection of average PCO2 in the first 3 days) was calculated using the formula described by Fabres et al.11 To calculate the time-weighted PCO2, we multiplied the PCO2 value with the time interval from birth for first gas/or from the previous blood gas in hours. These numbers were added together and divided by the postnatal age at last blood gas or 72 hours, whichever was earlier. (3) Cumulative hypercapnia index to hypercapnia was modified from previous definitions and with the consideration of both the duration and severity of hypercapnia. Whenever PCO2 was 65 mm Hg, the difference (PCO2 65) was multiplied by the duration of hypercapnia taken as the interval between the blood gas 65 and the preceding blood gas.12 All these values were added to calculate the cumulative index of hypercapnia exposure. This was divided by the age of patient in hours when last blood gas was obtained to yield the cumulative hypercapnia index per hour of life. Hypocapnia was defined as any PCO2 < 35 mm Hg. Normocapnia was defined as maintenance of all PCO2 values between 35 and 64 mm Hg in the first 3 days of life. Fluctuation in PCO2 was defined as the maximum difference between two consecutive PCO2 values within a 6-hour interval for each infant. We chose this short interval of 6 hours empirically to evaluate relatively quick fluctuations of PCO2 value in a patient. To evaluate the response of health care professionals to severe hypocapnia (blood gas with PCO2 < 30 mm Hg), we collected American Journal of Perinatology
Vol. 32
No. 9/2015
information on both the timing of attempted change on the ventilator, as well as time to reach PCO2 > 30 mm Hg. Patients were assigned the maximum form of respiratory support they received in the first 3 days of life. Occurrence of severe IVH was defined as grade III or IV IVH.13 In our center, preterm infants are resuscitated in the delivery room using blended oxygen, starting at 30% and adjusting based on the pulse oximeter readings. A T-piece resuscitator is available for delivery of CPAP or positive pressure ventilation. The initial dose of surfactant (Survanta [AbbVie Inc., North Chicago, IL]) is administered only for infants who are intubated and mechanically ventilated. Subsequent doses are administered based on the need for supplemental oxygen and high mean airway pressure on the ventilator. Ventilator adjustments are based on blood gases, which are typically performed every 4 to 6 hours. We use a permissive hypercarbia strategy, allowing PCO2 to be in the 45 to 64 mm Hg range as long as pH is > 7.20.
Statistical Analysis Descriptive statistical analyses were performed using the SPSS software version 19.0 (SPSS Inc., Chicago, IL). Results were expressed as mean ( standard deviation), median (range), or n (%) as appropriate. Bivariate comparisons of hypercapnia measures and fluctuations were performed using the t-test for continuous variables or Mann–Whitney U test as appropriate and chi-square test for categorical variables. Binary logistic regression analysis was performed to examine the associations between fluctuation in PCO2, and severe IVH while controlling for gestational age, ANS exposure, gender, Apgar score at 5 minutes, and hypercapnia. Higher gestational age and ANS exposure have previously been noted to be associated with the reduction in IVH.14 Apgar score at 5 minutes and the presence of hypercapnia (as a marker of respiratory severity) were also controlled for in the regression model. A p-value < 0.05 was considered significant.
Results Study Population During the study period, there were 342 live births with birth weight < 1,250 g. A total of 37 patients received only comfort care or died in the delivery room. Overall 20 patients received only nasal cannula or no respiratory support. A total of 285 neonates were included in the analysis for description of blood gas parameters. Noninvasive respiratory support, including CPAP or nasal intermittent positive pressure ventilation was the maximum form of respiratory support in 46 (16%) patients, while 239 (84%) were intubated and received mechanical ventilation. ►Table 1 describes the baseline characteristics and measures of hypercapnia, hypocapnia, and fluctuations in PCO2 within 72 hours in the study population. On initial blood gas, rates of hypocapnia, hypercapnia, and normocapnia were 14.4, 13, and 72.6%, respectively. Fluctuation in PCO2 10 was seen in 30.2%, 11 to 20 in 39.6%, and > 20 in 30.2%. The median time to wean the ventilator support following hypocapnia on the blood gas was 20 minutes. The median time for PCO2 to reach above 30 mm Hg was 4 hours.
Downloaded by: National University of Singapore. Copyrighted material.
840
Intraventricular Hemorrhage in Premature
Altaany et al.
841
Table 1 Baseline characteristics and ventilation profile
a
Gestational age (wk)
27 (25–28)
Birth weight (g)
890 (706–1,090)
Race (African American)
223 (82%)
Males
134 (47%)
Complete course of antenatal steroids
178 (62%)
Vaginal delivery
92 (32%)
Histologic chorioamnionitis
119 (46%)
Apgar score at 5 min < 5
50 (18%)
Delivery room intubation
183 (66%)
Time of first gas (h)
1 (1–4)
Number of gases in first 72 h
14 (11–16)
Exclusive hypocapnia (PCO2 < 35)
145 (51%)
Exclusive hypercapnia (PCO2 65)
25 (9%)
Normocapnia
57 (20%)
Any hypocapnia
195 (68%)
Any hypercapnia
75 (26%)
Both hypercapnia and hypocapnia
58 (20%)
Time weighted PCO2 mm Hg
40 (37–45)
Cumulative hypercapnia index in first 72 h of life
36 (0–99)
Cumulative hypercapnia index per h of life
0 (0–0.2)
Highest fluctuation in PCO2 mm Hg
14 (10–22)
Median (interquartile range) or n (%).
A total of 43 (15%) patients died of whom 17 did not have head ultrasound before death and were excluded from further analysis, leaving total of 268 patients in the analysis for severe IVH. Severe IVH (grade 3 or 4) was diagnosed in 41 (15%). Among patients, who had large fluctuation in PCO2, the rate of severe IVH was 23%, compared with 11% in patients, without large fluctuation in PCO2 (p ¼ 0.02). We further explored the association of severe IVH with known factors including gestation, gender, ANSs exposure, Apgar score at 5 minutes, and PCO2 derangements in a bivariate model (►Table 2). Low gestational age, lack of complete course of ANS, male gender, low Apgar score, highest PCO2, and maximum difference in PCO2 were significantly different in groups with and without severe IVH. A binary logistic regression analysis for severe IVH, controlling for gestational age, gender, complete course of ANS, Apgar score at 5 minutes (< 5), and hypercapnia (as a surrogate for severity of respiratory illness) noted that male gender, ANSs, and maximum difference in PCO2 were associated with severe IVH (►Table 3).
Discussion This single center study of a recent cohort of extreme premature infants requiring respiratory support provides some insight on the current clinical practices in PCO2 targeting.
The finding that only 20% of infants were in the target carbon dioxide range is consistent with one previous study noting poor compliance with the target oxygen saturations.15 The nurse–patient ratio, extent of respiratory support, and postmenstrual age were factors associated with achievement of target oxygen saturation range. We speculate that achievement of target carbon dioxide levels is potentially modifiable by some of these factors. We noted a high incidence of both hypercapnia and hypocapnia, which often coexisted. Hypocapnia was encountered in over 60% of the patients at one point in the first 3 days of life. Although, hypocapnia was not associated in this study or from previous reports with severe IVH, hypocapnia is known to be related to poor neurodevelopmental outcomes.12,16 This may be related to cerebral vasoconstriction in response to low PCO2 with subsequent cerebral ischemia.10 As hypocapnia is iatrogenic in most cases, it may be amenable to improvement with change in current practices. We examined various measures of hypercapnia in relation to severe IVH. Time weighted PCO2 was not noted to be associated with severe IVH. This supports earlier findings of Fabres et al. We speculate this due to a neutralizing effect of hypocapnia and hypercapnia with their frequent coexistence in the same subject (►Table 2). We then explored the relationship of the cumulative index of hypercapnia exposure and severe IVH. This again was also American Journal of Perinatology
Vol. 32
No. 9/2015
Downloaded by: National University of Singapore. Copyrighted material.
Variablea
Intraventricular Hemorrhage in Premature
Altaany et al.
Table 2 Factors associated with severe intracranial hemorrhage (grade III or IV) Severe ICH present n ¼ 41
Variable
a
Severe ICH absent n ¼ 227
p-Value
Gestational age (wk)
25.6 1.9
27 2.5
< 0.01
Birth weight (g)
862 223
899 220
0.34
Complete antenatal steroids course
13 (32%)
158 (70%)
< 0.01
Vaginal delivery
15 (37%)
73 (32%)
0.58
Male gender
26 (63%)
98 (43%)
0.02
Histologic chorioamnionitis
18 (56%)
98 (46%)
0.34
Apgar < 5 at 5 min
14 (37%)
31 (13.7%)
< 0.01
Highest PCO2
64.2 16.1
57.9 16
0.02
Any hypercapnia 65
14 (34%)
49 (21.5%)
0.11
Max difference in PCO2 (mm Hg)
24 17.1
16 12.6
< 0.01
Maximum difference > 10
35 (85%)
146 (65%)
0.07
Maximum difference > 20
20 (49%)
54 (24%)
0.02
Time weighted PCO2 mm Hg
40.7 4.8
40.7 6.3
0.88
Cumulative hypercapnia index mm Hg per h
0 (0–485)
0 (0–462)
0.08
Cumulative hypercapnia index per h of life
0 (0–6.8)
0 (0–6.4)
0.076
Mean (SD), median (range) or n (%).
Table 3 Logistic regression model for severe intracranial hemorrhage Variable
Odds ratio
95% confidence interval
p-Value
Gestation
0.83
0.69–1.01
0.05
Complete ANS
0.22
0.1–0.48
< 0.01
Male gender
2.5
1.2–5.4
0.015
5 minute Apgar score < 5
1.15
0.43–3.03
0.79
Hypercapnia
1.24
0.43–3.6
0.69
1.3
1.03–1.6
0.033
Max difference in PCO2 a
a
For every difference of 10 mm Hg between consecutive gases.
not significantly different in patients with and without severe IVH. Duration of hypercapnia may not be a significant factor associated with severe IVH. Hypercapnia and its association with severe IVH have been previously noted in few studies.9,17–20 Our study also confirmed this association. Fluctuations in PCO2 have also been noted to affect CBF with loss of cerebral autoregulation and increased risk of IVH11 and poor neurodevelopmental outcomes.21 In the current study, we explored the effect of fluctuation in PCO2 within a short interval on the risk of severe IVH. We chose a maximum of 6 hours interval between two consecutive blood gases to study effect of relatively acute fluctuations in PCO2. Our current neonatal intensive care unit protocol is to obtain blood gases every 4 to 6 hours in the first 3 days of life in extreme premature infants. To the best of our knowledge, this is the first study that describes the role of fluctuation in PCO2 within such short interval and its relation to severe IVH. Fabres et al in an earlier report have described the effect of PCO2 fluctuations and severe IVH. This was described using the magnitude of standard deviation around American Journal of Perinatology
Vol. 32
No. 9/2015
the mean PCO2 levels as well as the absolute difference between highest and lowest PCO2 levels at any point in the first 96 hour as markers of fluctuation. In our model, however, we examined the effect of PCO2 dispersion more acutely, by assessing fluctuation in PCO2 within 6 hours at most. After adjustment for gestational age, gender, ANS, and hypercapnia, fluctuation in PCO2 of 10 mm Hg in the first 3 days of life was associated with an increase in the risk of severe IVH (odds ratio 1.3). Interestingly, hypercapnia lost significance in this model. This may imply that fluctuations in PCO2 may be more significantly associated with severe IVH than the mere presence of hypercapnia. A major limitation in our study was the timing of blood gases. Frequency of obtaining blood gases was not standardized among all patients. The timing of the blood gases was determined by the clinical team, although our usual practice was to obtain blood gases every 4 to 6 hours in first 3 days of life. Sicker infants likely had more blood gases obtained, compared with neonates on lower respiratory support. Rates of hypercapnia and degree of fluctuation may be different
Downloaded by: National University of Singapore. Copyrighted material.
842
when PCO2 is monitored continuously. Arterial blood was obtained for analysis in 167 (58%) of the cohort. Arterialized capillary samples were used in absence of arterial lines. Due to this observational design of the study, there may be other unknown confounders associated with hypercapnia/fluctuations in PCO2 which may also predispose to severe IVH. A randomized controlled trial to study the effects of hypercapnia and rapid fluctuation in PCO2 in humans is not ethical.
7
8 9
Conclusion Normocapnia in the first 3 days of life in premature infants needing respiratory support was uncommon in our patient cohort with high rates of both hypercapnia and hypocapnia. Fluctuation in PCO2 was significantly associated with severe IVH. Continuous forms of PCO2 monitoring, such as transcutaneous PCO2 or end tidal CO2 monitoring may help reduce the incidence of these derangements and improve outcomes. Future studies using continuous forms of PCO2 monitoring may result in a better understanding of the role of hypercapnia and fluctuations in PCO2 in the development of severe IVH.
10
11
12
13
Conflict of Interest The authors declare no conflict of interest. 14
References
15
1 Futagi Y, Toribe Y, Ogawa K, Suzuki Y. Neurodevelopmental
2
3
4
5
6
outcome in children with intraventricular hemorrhage. Pediatr Neurol 2006;34(3):219–224 Bassan H, Feldman HA, Limperopoulos C, et al. Periventricular hemorrhagic infarction: risk factors and neonatal outcome. Pediatr Neurol 2006;35(2):85–92 Sherlock RL, Anderson PJ, Doyle LW; Victorian Infant Collaborative Study Group. Neurodevelopmental sequelae of intraventricular haemorrhage at 8 years of age in a regional cohort of ELBW/very preterm infants. Early Hum Dev 2005;81(11):909–916 Pryds O, Greisen G. Effect of PaCO2 and haemoglobin concentration on day to day variation of CBF in preterm neonates. Acta Paediatr Scand Suppl 1989;360:33–36 Hansen NB, Brubakk AM, Bratlid D, Oh W, Stonestreet BS. The effects of variations in PaCO2 on brain blood flow and cardiac output in the newborn piglet. Pediatr Res 1984;18(11): 1132–1136 Perlman JM, Goodman S, Kreusser KL, Volpe JJ. Reduction in intraventricular hemorrhage by elimination of fluctuating cere-
16
17
18
19 20 21
Altaany et al.
bral blood-flow velocity in preterm infants with respiratory distress syndrome. N Engl J Med 1985;312(21):1353–1357 Van Bel F, Van de Bor M, Stijnen T, Baan J, Ruys JH. Aetiological rôle of cerebral blood-flow alterations in development and extension of peri-intraventricular haemorrhage. Dev Med Child Neurol 1987;29(5):601–614 Ballabh P. Intraventricular hemorrhage in premature infants: mechanism of disease. Pediatr Res 2010;67(1):1–8 Kaiser JR, Gauss CH, Pont MM, Williams DK. Hypercapnia during the first 3 days of life is associated with severe intraventricular hemorrhage in very low birth weight infants. J Perinatol 2006; 26(5):279–285 Wyatt JS, Edwards AD, Cope M, et al. Response of cerebral blood volume to changes in arterial carbon dioxide tension in preterm and term infants. Pediatr Res 1991;29(6):553–557 Fabres J, Carlo WA, Phillips V, Howard G, Ambalavanan N. Both extremes of arterial carbon dioxide pressure and the magnitude of fluctuations in arterial carbon dioxide pressure are associated with severe intraventricular hemorrhage in preterm infants. Pediatrics 2007;119(2):299–305 Shankaran S, Langer JC, Kazzi SN, Laptook AR, Walsh M; National Institute of Child Health and Human Development Neonatal Research Network. Cumulative index of exposure to hypocarbia and hyperoxia as risk factors for periventricular leukomalacia in low birth weight infants. Pediatrics 2006;118(4):1654–1659 Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978; 92(4):529–534 Chawla S, Natarajan G, Rane S, Thomas R, Cortez J, Lua J. Outcomes of extremely low birth weight infants with varying doses and intervals of antenatal steroid exposure. J Perinat Med 2010;38(4):419–423 Sink DW, Hope SA, Hagadorn JI. Nurse:patient ratio and achievement of oxygen saturation goals in premature infants. Arch Dis Child Fetal Neonatal Ed 2011;96(2):F93–F98 Collins MP, Lorenz JM, Jetton JR, Paneth N. Hypocapnia and other ventilation-related risk factors for cerebral palsy in low birth weight infants. Pediatr Res 2001;50(6):712–719 Vela-Huerta MM, Amador-Licona M, Medina-Ovando N, AldanaValenzuela C. Factors associated with early severe intraventricular haemorrhage in very low birth weight infants. Neuropediatrics 2009;40(5):224–227 Thome UH, Ambalavanan N. Permissive hypercapnia to decrease lung injury in ventilated preterm neonates. Semin Fetal Neonatal Med 2009;14(1):21–27 Zhou W, Liu W. Hypercapnia and hypocapnia in neonates. World J Pediatr 2008;4(3):192–196 Fritz KI, Delivoria-Papadopoulos M. Mechanisms of injury to the newborn brain. Clin Perinatol 2006;33(3):573–591, v v McKee LA, Fabres J, Howard G, Peralta-Carcelen M, Carlo WA, Ambalavanan N. PaCO2 and neurodevelopment in extremely low birth weight infants. J Pediatr 2009;155(2):217–21.e1
American Journal of Perinatology
Vol. 32
No. 9/2015
843
Downloaded by: National University of Singapore. Copyrighted material.
Intraventricular Hemorrhage in Premature
