This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Claure, N. Search for Related Content PubMed PubMed Citation Articles by Claure, N.

---

REVIEW

Automated Regulation of Inspired Oxygen in Preterm Infants: Oxygenation Stability and Clinician Workload

From the Division of Neonatology, Department of Pediatrics, University of Miami Miller School of Medicine, Miami, Florida.

Address correspondence and reprints requests to Nelson Claure, Division of Neonatology, Department of Pediatrics, University of Miami School of Medicine, P.O. Box 016960 R-131, Miami, FL 33101. Address e-mail to NClaure{at}miami.edu.

Abstract

Premature infants are at an increased risk of ophthalmic, neurologic, and respiratory sequelae related to inadequate maintenance of oxygenation and exposure to increased levels of inspired oxygen. Management of inspired oxygen is complicated in this population by an increased variability in oxygenation. Automated regulation of the fraction of inspired oxygen is a technology that has a potential of improving such outcomes as well as impacting personnel workload. This is a review of current experimental evidence on the effectiveness of automated regulation of inspired oxygen and its effects on oxygenation variability and personnel workload during the care of premature infants.

Supplemental oxygen is given to otherwise hypoxemic premature infants to maintain adequate oxygenation and to avoid the deleterious effects of hypoxia. However, delivery of supplemental oxygen is not a risk-free therapy. Because of their premature birth, these infants are often exposed to oxygen at a time when their lungs, antioxidant system, and retina are too immature. Also, these infants often require supplemental oxygen for several days or weeks after birth. Therefore, they are at an increased risk for oxidative stress, lung injury, and retinopathy of prematurity (1–6). Management of the supplemental oxygen aims at minimizing exposure to increased levels of inspired oxygen while maintaining adequate oxygenation.

OXYGENATION VARIABILITY IN PREMATURE INFANTS

Under routine clinical conditions, management of the fraction of inspired oxygen (Fio₂) is often complicated by fluctuations in oxygenation. These fluctuations vary in severity and frequency. Fluctuations associated with worsening or improvement of a respiratory condition occur gradually, and are relatively simple to correct while other fluctuations can occur quite frequently and are often characterized by rapid and severe changes in oxygenation (7–11).

Compared to other patient populations in intensive care, premature infants are at increased risk of developing complications related to fluctuations in oxygenation, which are believed to influence the development of retinopathy of prematurity (12–17). Intermittent hypoxemia can also have a negative impact on airways, lung vasculature, and other organs (18–21) while exposure to high concentrations of inspired oxygen can lead to development of permanent respiratory sequelae (1–4), which is associated with impaired neurologic outcome (22,23).

These fluctuations in oxygenation are detected by arterial oxygen saturation monitors (Spo₂) and trigger a personnel response. The response to the occurrence of hypoxemia usually consists of a transient increase in Fio₂ until normoxemia is restored. In the event of hyperoxemia, Fio₂ is gradually reduced. Under standard clinical conditions, personnel-to-patient ratio and increasing amounts of caregiver workload often do not permit full attention to these tasks. As a result, the response time when Spo₂ is outside an intended range of oxygenation is affected. Delayed responses can prolong exposure to unnecessarily high concentrations of supplemental oxygen or periods of hypoxemia.

Recent, multicenter data has shown that premature infants who require supplemental oxygen spend only about half of the time within individual center’s intended range of oxygenation. Of the remaining time, about 20% is spent with Spo₂ readings below and about 30% above the intended range (24,25). Spo₂ readings in hyperoxemia in infants receiving supplemental oxygen could only be attained by providing unnecessarily high concentrations of inspired oxygen.

RATIONALE FOR AUTOMATED REGULATION INSPIRED OXYGEN

Automated regulation of the inspired oxygen in premature infants is intended to address the above-mentioned issues by reducing periods of hyper- and hypoxemia, and limiting the exposure to high concentrations of inspired oxygen. The rationale that supports the development of automatic Fio₂ control is based on the timely Fio₂ adjustment to increase the alveolar oxygen concentration, followed by prompt weaning as soon as the additional oxygen is no longer needed.

These tasks could also be achieved with manual adjustments by clinical personnel. However, they can become time-consuming and increase workload.

EXPERIENCE WITH AUTOMATED Fio₂ REGULATION: OXYGENATION STABILITY AND WORKLOAD

Various studies have shown the feasibility of automated Fio₂ regulation for the care of premature infants. Moreover, automated Fio₂ regulation shows improvement in oxygenation stability around intended ranges of oxygenation compared to routine care and even when compared to nursing care dedicated to oxygen regulation (26–32).

When assessing the comparisons between automated and manual Fio₂ regulation in regard to oxygenation stability and frequency of manual interventions, it is important to consider the basal rate of variability of the studied population and personnel responsiveness. Comparisons of automated Fio₂ regulation to manual interventions are not likely to benefit infants who present with minimal or no variability in oxygenation. On the other hand, comparing automated Fio₂ regulation to routine care that is relatively unresponsive to fluctuations in oxygenation is likely to show greater benefits in comparison to more dedicated or attentive care. Also, depending on predetermined objectives for the automated system, not all such systems may be designed to address the different forms of oxygenation variability.

Clinically, the most important goal of automated Fio₂ regulation is to stabilize oxygenation. Automated Fio₂ regulation prolonged the time within an intended range of oxygenation compared to routine care consisting of adjustments at two hourly intervals (26,27,29) or manual adjustments every 20–30 min, even when compared to a more dedicated care with adjustments every 2–3 min (28). Moreover, automated Fio₂ prolonged the time within an intended range of oxygenation even when compared to fully dedicated care (29,32). These data illustrate the efficacy of this type of automation.

Infants with very frequent and/or acute fluctuations in oxygenation increase the demand for personnel time and also are a greater challenge for automated Fio₂ regulation. Not all systems described in the literature have been developed to address these fluctuations. In some of these, the response of the automated system consists of a user alert (26) or, in others require manual intervention because of inadequate system response (27,30). An automated Fio₂ controller designed to assist during acute episodes of hypoxemia increased the time within an intended range in comparison to the time spent by a fully dedicated nurse at bedside in a group of preterm infants who presented with an average of 15 episodes of hypoxemia per hour (32).

Under these challenging conditions, quantification of the effort required for tight control showed that the nurse adjusted Fio₂ a mean of 29 times an hour. This frequency does not fully indicate workload since continuous dedicated attention is needed at the onset, throughout, and at the end of the episode of hypoxemia. Reducing the workload resulting from these repetitive tasks may lead to a more effective use of personnel.

ASSESSING OXYGENATION

Perhaps, the most important component in the routine or automated process of Fio₂ regulation is continuous assessment of oxygenation. This can be done using indwelling Pao₂ electrodes inserted through an umbilical artery catheter, transcutaneous PO₂ electrodes (tcPO₂) or arterial oxygen hemoglobin saturation measurements by pulse oximetry (Spo₂). Indwelling measurements are limited by availability of an invasive line. tcPo₂ is noninvasive, but depends on electrode temperature for accuracy. The continuity of tcPo₂ measurements is therefore affected by changes in application site to avoid thermal injury. Spo₂ provides a continuous noninvasive assessment of oxygenation. Spo₂ setup is relatively simple and is widely used in newborn intensive care units.

PULSE OXIMETRY TO DETECT HYPEROXEMIA AND HYPOXEMIA

Interpretation of Spo₂ to detect and respond to hypo- or hyperoxemia is done in the context of the relationship between Spo₂ and Pao₂, as depicted by the sigmoid-shaped O₂ dissociation curve. In the range of hyperoxemia, relatively small changes in Spo₂ are associated with large changes in Pao₂. Data (33) show that a threshold Spo₂ around 94% and 96% would correctly classify as hyperoxemic most Pao₂ readings above 80 or 90 mm Hg. Most false-positive readings would be in the high–normal range (>60 mm Hg) and none in the low (<40 mm Hg) or low–normal Pao₂ range (40–60 mm Hg). Therefore, a gradual reduction in Fio₂ in response to high Spo₂ will correct most hyperoxemic events and is unlikely to lead to hypoxemia.

In the range of hypoxemia, most readings below a Spo₂ threshold around 85% and 88% are associated with Pao₂ readings in hypoxemia (<40 mm Hg), a small fraction with readings in the low–normal range (40–60 mm Hg) and none in the high Pao₂ range (>80 mm Hg). Therefore, a Fio₂ increase in response to hypoxemia detected by Spo₂ will be appropriate in most instances and unlikely to result in hyperoxemia.

ARTIFACT OR TRUE HYPOXEMIA?

The reliability of Spo₂ in premature infants is especially sensitive to motion because of the relatively low pulse pressure. Motion can produce venous blood pulsation and/or disrupt the optical pathway from the transmitter to receiver side of the probe. Despite documented improvements in Spo₂ technology, some hypoxemic episodes detected by pulse oximetry are considered artifactual. Visual inspection of the infant may reveal significant movement of extremities, absence of conclusive signs of cyanosis, and a transient mismatch between oximeter pulse rate and heart rate monitors. However, these observations cannot exclude true hypoxemia. To the contrary, increased infant activity and changes in heart rate, lung volume, and ventilation have been associated with hypoxemia episodes (9,10). Moreover, hypoxemia is more prevalent during arousal and indeterminate sleep states compared to periods of active or quiet sleep (11).

The response to an episode of hypoxemia and simultaneous motion, either by an automated system or by a caregiver, involves the risk of unnecessary oxygen exposure when the episode is artifactual. While the value of avoiding exposure to unnecessary supplemental oxygen is unquestionable, failure to assist a true hypoxemic episode because suspicion of artifact may allow alveolar and tissue hypoxia to increase.

Accuracy of Spo₂ is influenced by intrinsic patient conditions, such as low perfusion or inadequate setup of the oximeter (34). These conditions, however, are particular not only to automated regulation of Fio₂ but also to routine care. It is ultimately the clinician who decides if the use of pulse oximetry is appropriate and sufficiently reliable to monitor an infant.

When used in infants with frequent episodes of hypoxemia, automated Fio₂ regulation during every episode resulted in negligible overshoot with <3% of hypoxemia episodes followed by readings in hyperoxemia (32). If most of these episodes of hypoxemia had been artifactual, the additional oxygen would have had a higher rate of overshoot.

SIGNAL DROP OUT: MISSING Spo₂ INFORMATION

Poor signal quality can lead to periods of missing Spo₂ information, as determined by the built-in oximeter validation algorithms. This leaves clinical personnel or an automatic system without feedback information. As mentioned above, hypoxemia can be associated with body movement; therefore, there is the possibility that a period of missing Spo₂ information caused movement is accompanied by undetected hypoxemia. On the other hand, if the missing Spo₂ period is caused by technical factors, it could result in a period of unnecessary oxygen exposure. This, however, may resolve as soon as the clinician responds to the appropriate alarms and warnings given by the pulse oximeter and/or the automatic system. Data obtained during clinical use of automated Fio₂ control indicated that most periods of missing Spo₂ data were followed by periods of hypoxemia (32).

AUTOMATED SYSTEM RESPONSE

The method of automated Fio₂ regulation should modulate its response to severity, duration, rate of change and direction of the fluctuation in oxygenation. Optimization of the timing between Spo₂ events and Fio₂ adjustments is important when the objective is to assist rapidly ensuing and frequent hypoxemia episodes, while requirements are less for assisting infants who present with mild or gradual changes in oxygenation. A long delay may result in prolongation of hypoxemia or exposure to higher concentration of inspired oxygen when it is no longer needed.

Spo₂ data are averaged over a running window to reduce the effect of short variations. A long averaging window can slow detection of changes in oxygenation, and can increase the inertia of automated control leading to overshoot. A shorter averaging window enables the automated system to determine its own response time based on the variables mentioned above.

The mode of delivery should be optimized to produce changes in the inspired gas with minimal delays in setting adjustments at the controller. Optimized gas mixing can be accomplished by increased circulating flow rates in the ventilator and minimizing reservoir spaces.

RISKS AND BENEFITS

Failure or suboptimal function of the measurement, delivery or control components of automated Fio₂ control may involve risks of exposure to hypoxemia or unnecessarily high concentrations of inspired oxygen. The use of this technology may result in only a relative increase of these risks, which may be, in part, present during routine care because with the exception of the manual regulation component, the measurement and delivery components are standard.

Monitoring and assistance during routine care may be sufficient for infants with mild or gradual fluctuations in oxygenation or those whose disease condition requires high Fio₂ continuously. On the other hand, infants who present with frequent and acute changes in oxygenation may obtain a greater benefit from automated assistance. Dedicated assistance to these infants imposes additional workload on personnel.

Automated Fio₂ regulation is aimed mainly at conducting repetitive and time consuming tasks. It is meant to be used in a manner in which the time dedicated by the caregiver to patient care is not affected by these tasks. However, automated Fio₂ regulation may in some instances result in an unwanted reduction in the time dedicated by the caregiver to patient monitoring. Since hypoxemia episodes in preterm infants have a wide etiology, automated Fio₂ regulation could mask the effects of a triggering event such as hypoventilation. Proper and timely monitoring of cardiorespiratory variables should be assured to avert this type of situation. On the other hand, one could argue that an automated Fio₂ increase should at least reduce the detrimental effects until corrective actions are taken.

Assistance for every episode of hypoxemia with an increase in Fio₂ could result in additional exposure to oxygen. Timely weaning when hypoxemia resolves should limit the exposure to oxygen. As mentioned above, overshoot (hyperoxemia) was observed in <3% of hypoxemia episodes corrected with an automated Fio₂ increase. Nonetheless, compared to no intervention in the form of supplemental oxygen, the lungs would be exposed to higher concentrations of oxygen and therefore an increased risk of lung injury. A predetermined absence of intervention during hypoxemia episodes assumes that the effects of these episodes on the central nervous system and other organs are irrelevant to outcome.

SUMMARY

Supplemental oxygen administration, although a necessary and often life-saving therapy, has side effects that are of relevance in the premature infant population. This population has a high risk for respiratory, neurologic, and ophthalmic compromise. Therefore, there is a need for gentler, yet effective, oxygenation support.

The emergent availability of improved technology for oxygenation monitoring and computing has facilitated the development of automated modes of regulation of supplemental oxygen for the preterm infant. Although this technology has been only used experimentally, preliminary results suggest future impact on patient care and practice. This, however, remains to be tested. Future research will determine the role of this strategy in improving pulmonary, ophthalmic, and neurological outcome in premature infants, as well as in determining the true impact on caregiver workload.

Footnotes

Accepted for publication April 19, 2007.

REFERENCES

1. Taghizadeh A, Reynolds EO. Pathogenesis of bronchopulmonary dysplasia following hyaline membrane disease. Am J Pathol 1976;82:241–64[Abstract]
2. Palta M, Gabbert D, Weinstein MR, Peters ME. Multivariate assessment of traditional risk factors for chronic lung disease in very low birth weight neonates. J Pediatr 1991;119:285–92[Web of Science][Medline]
3. Saugstad OD. Chronic lung disease: the role of oxidative stress. Biol Neonate 1998;74:21–8[Web of Science][Medline]
4. Saugstad OD. Free radical disease in neonatology. Semin Neonatol 1998;3:22–38
5. Frank L, Sosenko IRS. Failure of premature rabbits to increase antioxidant enzymes during hyperoxic exposure: increased susceptibility to pulmonary oxygen toxicity compared with term rabbits. Pediatr Res 1991;29:292–6[Web of Science][Medline]
6. Avery G, Glass P. Retinopathy of prematurity: what causes it? Clin Perinatol 1988;15:917–28[Web of Science][Medline]
7. Garg M, Kurzner SI, Bautista DB, Keens TG. Clinically unsuspected hypoxia during sleep and feeding in infants with bronchopulmonary dysplasia. Pediatrics 1988;81:635–42[Abstract/Free Full Text]
8. Durand M, McEvoy C, MacDonald K. Spontaneous desaturations in intubated very low birth weight infants with acute and chronic lung disease. Pediatr Pulmonol 1992;13:136–42[Web of Science][Medline]
9. Bolivar JM, Gerhardt T, Gonzalez A, Hummler H, Claure N, Everett R, Bancalari E. Mechanisms for episodes of hypoxemia in preterm infants undergoing mechanical ventilation. J Pediatr 1995;127:767–73[Web of Science][Medline]
10. Dimaguila MA, DiFiore JA, Martin R, Miller MJ. Characteristics of hypoxemic episodes in very low birth weight infants on ventilatory support. J Pediatr 1997;130:577–83[Web of Science][Medline]
11. Lehtonen L, Johnson MW, Bakdash T, Martin RJ, Miller MJ, Scher M. Relation of sleep state to hypoxemic episodes in ventilated extremely-low-birth-weight infants. J Pediatr 2002;141:363–9[Web of Science][Medline]
12. Penn JS, Henry MM, Tolman BL. Exposure to alternating hypoxia and hyperoxia causes severe proliferative retinopathy in the newborn rat. Pediatr Res 1994;36:724–31[Web of Science][Medline]
13. Saito Y, Omoto T, Cho Y, Hatsukawa Y, Fujimura M, Takeuchi T. The progression of retinopathy of prematurity and fluctuation in blood gas tension. Albrecht Von Graefes Arch Ophthalmol 1993;231:151–6
14. Reynaud X, Dorey CK. Extraretinal neovascularization induced by hypoxic episodes in the neonatal rat. Invest Ophthalmol Vis Sci 1994;35:3169–77[Abstract/Free Full Text]
15. Phelps DL, Rosenbaum A. Effects of marginal hypoxemia on recovery from oxygen-induced retinopathy in the kitten model. Pediatrics 1984;73:1–6[Abstract/Free Full Text]
16. McColm JR, Cunningham S, Wade J, Sedowofia K, Gellen B, Sharma T, McIntosh N, Fleck BW. Hypoxic oxygen fluctuations produce less severe retinopathy than hyperoxic fluctuations in a rat model of retinopathy of prematurity. Pediatr Res 2004;55:107–13[Web of Science][Medline]
17. Cunningham S, McColm JR, Wade J, Sedowofia K, McIntosh N, Fleck B. A novel model of retinopathy of prematurity simulating preterm oxygen variability in the rat. Invest Ophthalmol Vis Sci 2000;41:4275–80[Abstract/Free Full Text]
18. Tay-Uyboco JS, Kwiatkowski K, Cates DB, Kavanagh L, Rigatto H. Hypoxic airway constriction in infants of very low birth weight recovering from moderate to severe bronchopulmonary dysplasia. J Pediatr 1989;115:456–9[Web of Science][Medline]
19. Unger M, Atkins M, Briscoe WA, King TK. Potentiation of pulmonary vasoconstrictor response with repeated intermittent hypoxia. J Appl Physiol 1977;43:662–7[Abstract/Free Full Text]
20. Custer JR, Hales CA. Influence of alveolar oxygen on pulmonary vasoconstriction in newborn lambs versus sheep. Am Rev Respir Dis 1985;132:326–31[Web of Science][Medline]
21. Barlow B, Santulli T. Importance of multiple episodes of hypoxia or cold stress on the development of enterocolitis in an animal model. Surgery 1975;77:687–90[Web of Science][Medline]
22. Lifschitz MH, Seilheimer DK, Wilson GS, Williamson WD, Thurber SA, Desmond MM. Neurodevelopmental status of low birth weight infants with bronchopulmonary dysplasia requiring prolonged oxygen supplementation. J Perinatol 1987;7:127–32[Medline]
23. Schmidt B, Asztalos EV, Roberts RS, Robertson CM, Sauve RS, Whitfield MF. Impact of bronchopulmonary dysplasia, brain injury, and severe retinopathy on the outcome of extremely low-birth-weight infants at 18 months: results from the trial of indomethacin prophylaxis in preterms. JAMA 2003;289:1124–9[Abstract/Free Full Text]
24. Laptook AR, Salhab W, Allen J, Saha S, Walsh M. Pulse oximetry in very low birth weight infants: can oxygen saturation be maintained in the desired range? J Perinatol 2006;26:337–41[Medline]
25. Hagadorn JI, Furey AM, Nghiem TH, Schmid CH, Phelps DL, Pillers DA, Cole CH. AVIOx Study Group. Achieved versus intended pulse oximeter saturation in infants born less than 28 weeks’ gestation: the AVIOx study. Pediatrics 2006;118:1574–82[Abstract/Free Full Text]
26. Beddis JR, Collins P, Levy NM, Godfrey S, Silverman M. New Technique for servo-control of arterial oxygen tension in preterm infants. Arch Dis Child 1979;54:278–80[Abstract/Free Full Text]
27. Dugdale RE, Cameron RG, Lealman GT. Closed-loop control of the partial pressure of arterial oxygen in neonates. Clin Phys Physiol Meas 1988;9:291–305[Medline]
28. Bhutani VK, Taube JC, Antunes MJ, Delivoria-Papadopoulos M. Adaptive control of the inspired oxygen delivery to the neonate. Pediatr Pulmonol 1992;14:110–7[Web of Science][Medline]
29. Urschitz MS, Horn W, Seyfang A, Hallenberger A, Herberts T, Miksch S, Popow C, Muller-Hansen I, Poets CF. Automatic control of the inspired oxygen fraction in preterm infants: a randomized crossover trial. Am J Respir Crit Care Med 2004;170:1095–100[Abstract/Free Full Text]
30. Morozoff PE, Evans RW. Closed-loop control of SaO₂ in the neonate. Biomed Instrum Technol 1992;26:117–23[Medline]
31. Sun Y, Kohane IS, Start AR. Computer-assisted adjustment of inspired oxygen concentration improves control of oxygen saturation in newborn infants requiring mechanical ventilation. J Pediatr 1997;131:754–6[Web of Science][Medline]
32. Claure N, Gerhardt T, Everett R, Musante G, Herrera C, Bancalari E. Closed-loop controlled inspired oxygen concentration for mechanically ventilated very low birth weight infants with frequent episodes of hypoxemia. Pediatrics 2001;107:1120–4[Abstract/Free Full Text]
33. Hay WW, Brockway JM, Eyzaguirre M. Neonatal pulse oximetry: accuracy and reliability. Pediatrics 1989;83:717–22[Abstract/Free Full Text]
34. Bucher HU, Keel M, Wolf M, von Siebenthal K, Duc G. Artifactual pulse-oximetry estimation in neonates. Lancet 1994;343:1135–6[Web of Science][Medline]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Claure, N. Search for Related Content PubMed PubMed Citation Articles by Claure, N.