Research ArticleOriginal Research

Outcomes of Preterm Infants Who Experienced Unplanned Extubation

Vrinda Nair, Prakash Kannan Loganathan, Helena Smith and Mithilesh Kumar Lal
Respiratory Care October 2022, 67 (10) 1320-1326; DOI: https://doi.org/10.4187/respcare.10005

Abstract

BACKGROUND: Unplanned extubation (UE) is associated with adverse outcomes. The aim of the study was to compare the clinical outcomes in preterm infants who experienced at least one UE to those who did not experience any UE.

METHODS: The matched cohort study compared ventilated preterm infants < 32 weeks who experienced UE to those who did not experience any UE. The main outcomes were duration of mechanical ventilation after matching, duration of hospital stay, retinopathy of prematurity (ROP) requiring intervention, and bronchopulmonary dysplasia (BPD).

RESULTS: Forty-seven infants were included in each group. The groups were matched for mechanical ventilation duration before UE, birth gestation, and birthweight. The duration of mechanical ventilation after matching (adjusted odds ratio [aOR] 14.8 [11.2–18.4], P = <.001), the total length of stay in the hospital (aOR 16.4 [3.7–29.2], P = .01), and severe ROP (aOR 6.7 [1.7–27.0], P = .007) were significantly higher in infants who experienced UE. After adjusting for mechanical ventilation duration, UE was not associated with ROP or BPD. However, infants who spent longer time on mechanical ventilation had higher odds of developing ROP (aOR 1.1 [1.0–1.2], P = .004) and BPD (aOR 1.5 [1.1–2.1], P = .01). Sensitivity analysis including infants who had UE and managed on noninvasive respiratory support showed significant association between UE and the outcomes of duration of mechanical ventilation, hospital length of stay, ROP, and BPD.

CONCLUSIONS: Infants who experienced UE had higher odds of spending longer time on mechanical ventilation and spent significantly more days in the hospital.

Introduction

Mechanical ventilation is one of the most common interventions carried out in preterm infants cared for in the neonatal ICU (NICU). One of the adverse events associated with mechanical ventilation is unplanned extubation (UE). Neonates in particular are at increased risk1 of UE because of small facial size, use of uncuffed endotracheal tube (ETT), longer duration of ventilation, shorter trachea, and limited use of sedation.2 UE is an avoidable event, and implementation of quality improvement (QI) bundles could help to reduce the rates of UE.3

The rates of UE vary from center to center. Whereas it is well understood that UE is associated with immediate adverse event, still it remains underreported in the NICU. The immediate adverse events associated with UE in neonates and children include hypothermia, need for cardiopulmonary resuscitation,3,4 hypoxia, hypercarbia, and hypocarbia.5,6 Emergent re-intubation following UE may result in airway trauma7 and subglottic stenosis.8 There are studies in the literature showing adverse short- and long-term outcomes in pediatric and adult ICU primarily relating to prolonged hospital stay and ventilation duration.9 Recently, Hatch et al published data on adverse hospital outcomes and increased costs in preterm infant with UE.10

We hypothesized that preterm infants experiencing UE have worse hospital outcomes. The objective of the study was to compare the clinical hospital outcomes in preterm infants who experienced at least one UE to those who did not experience any UE.

QUICK LOOK

Current Knowledge

Unplanned extubation (UE) is one of the most common adverse events in the ICU. Neonates in particular are at high risk for UE, and quality improvement initiatives can help reduce the rates of UE.

What this paper contributes to the current knowledge

UE was associated with significantly longer duration of mechanical ventilation. Infants who experienced UE had significantly longer hospital stay. Duration of mechanical ventilation was associated with severe retinopathy of prematurity and bronchopulmonary dysplasia.

Methods

Study Design, Setting, and Population

This was a matched retrospective cohort study conducted in a level III NICU. All preterm infants < 32 weeks who exclusively received their entire duration of ventilation episodes in the study center between April 2016–March 2020 were considered eligible for the study. We excluded infants ≥ 32 weeks and those who received ventilation episodes in other centers. In our unit, all infants who are ventilated are intubated orally. Intubations are carried out by medical staff, and the trained bedside nurse helps with the fixation of the ETT. Ventilated infants are not routinely sedated and do not receive a routine paralytic agent. Our facility provides one-to-one nursing care for ventilated infants, per national recommendations. We defined UE as any unexpected removal of the ETT and included the removal of the ETT when deemed blocked or not in the trachea by the clinical team.3 A QI project was launched aimed at reducing UE rates in July 2017. Standardizing ETT fixation, continuous scrutiny of fixation through checks, 2-person technique for providing care, and adverse event reporting were a few of the important interventions of this QI project.3 No other respiratory changes were introduced during the study period. The study was approved by the Health Research Authority (20/HRA/4741), United Kingdom.

Study Matching

We chose the following variables for one-to-one matching: (1) mechanical ventilation duration before the first UE, (2) completed gestation age at birth, and (3) birthweight. We included the next matched no-UE subject after a UE subject. For instance, if an infant had 5 d of mechanical ventilation before first UE, we identified infants who did not have UE with at least 5 d of mechanical ventilation and further matching done with gestational age (± 1 week) and birthweight (± 100 g). The variable of mechanical ventilation duration before the first UE used for matching was similar to the study by Hatch et al.10

Outcomes

The main outcomes of interest were agreed a priori by authors: duration of mechanical ventilation after the first UE, total duration of mechanical ventilation, duration of hospital stay, severe retinopathy of prematurity (ROP; requiring intervention), and bronchopulmonary dysplasia (BPD) defined as oxygen or respiratory support need at 36 weeks postmenstrual age. We reported other outcomes including late onset sepsis (any blood culture positivity after 72 h of age) and periventricular leukomalacia as evident at 6-week cranial ultrasound.

We performed sensitivity analysis including infants who had UE and were subsequently managed on noninvasive respiratory support (n = 8). We also performed a subgroup analysis on infants who had more than one UE (multiple) and compared their outcomes with those who had only one UE.

Statistical Analysis

After matching, standardized mean/median difference was used to understand any imbalance in the 2 groups, and a value < ± 0.1 was considered as negligible difference.11 Categorical variables were presented as proportions and the continuous variables as median with interquartile range (IQR). Exact McNemar test was used in the analysis of paired dichotomous outcomes. For paired continuous outcomes, Wilcoxon signed-rank test was used. In multivariable analysis, we used linear mixed-effect regression for continuous outcomes and logistic regression for categorical outcomes. Covariates with standardized difference of > 0.1 and those likely to be independently affecting the outcome were considered for regression analysis. Kaplan-Meier graphs were constructed for duration of ventilation after matching and the total duration of hospital stay. For comparison of infants with multiple versus single UE (unpaired samples), univariate analysis was performed using Mann-Whitney test for continuous outcomes and chi-square test for categorical outcomes. A P value of < .05 was considered statistically significant. Statistical analysis was performed using both Stata/SE (version 16.1, StataCorp, College Station, Texas) and IBM SPSS Statistics (version 27, IBM, Armonk, New York).

Results

Two hundred and nineteen preterm infants (< 32 weeks gestation) received courses of invasive mechanical ventilation exclusively in the study center during the study period. Of these, 55 (25%) infants experienced at least one UE. There were 94 episodes of UE. Eight infants who experienced UE were managed on noninvasive ventilation following the UE. We finally had 47 matched pairs for analysis. Variables before and after matching along with their standardized differences are provided in Table 1. For most variables, there was no imbalance between the matched pairs; however, there was a higher proportion of male infants and mothers who received antenatal steroids in the UE group. The median chronological age of first UE in our study was 10 d (IQR 3.0–15.5). The mortality was similar in both groups. None of the mortality was attributable to UE event.

View this table:
Table 1.

Demographics of Infants Before and After Matching

Unadjusted and adjusted odds ratio for all the outcomes are reported in Table 2. On univariate analysis, the duration of mechanical ventilation after matching, total duration of mechanical ventilation, hospital length of stay, ROP requiring intervention, and BPD at 36 weeks postmenstrual age were statistically higher in the UE group. For regression analysis, UE, sex, antenatal steroids, and admission temperature were included as these variables were imbalanced after matching between the 2 groups. On multivariate analysis, the duration of mechanical ventilation after matching, total duration of mechanical ventilation, hospital length of stay, and severe ROP were statistically significant higher in UE group. BPD was higher in UE group (87.5%) as compared to no UE group (60.5%). However, this did not reach statistical significance. Since mechanical ventilation duration is associated with BPD and ROP, we analyzed odds of these morbidities by further adjusting for post-matching mechanical ventilation duration. In this model, duration of mechanical ventilation was significantly associated with BPD (1.5 [1.1–2.1], P = .01) and ROP (1.1 [1.0–1.2], P = .004), and UE was not.

View this table:
Table 2.

Hospital Outcomes

On performing sensitivity analysis (including infants who were managed on noninvasive ventilation following UE, n = 55), there was negligible difference between the 2 groups for the matching variables (standardized mean/median difference < ± 0.1) (Supplementary Table 1, see related supplementary materials at http://www.rcjournal.com). Sex, antenatal steroids, Apgar score at one minute, and admission temperature were not balanced after matching. Both univariate and regression analysis showed significant association with UE and all the primary outcomes including BPD (Table 3).

View this table:
Table 3.

Hospital Outcome Including All Infants with Unplanned Extubation (Sensitivity Analysis)

Figure 1 shows the Kaplan-Meier curves for duration of mechanical ventilation after matching and hospital length of stay. Eighteen infants (38.3%) experienced more than one UE (Table 2) (See related supplementary materials at http://www.rcjournal.com). The median hospital length of stay in d (112.0 [108.5–133.5] vs 92.5 [69.3–110.3], P = .003) and severe ROP (62.5% vs 29%, P = .038) were significantly higher in infants who experienced multiple UE when compared to infants who experienced only one UE.

Fig. 1.
Fig. 1.

A: Kaplan-Meier curve for duration of ventilation after matching. B: Kaplan-Meier curve for hospital stay.

Discussion

In this matched cohort study, ventilated infants who experienced at least one UE while in the NICU spent significantly longer duration on mechanical ventilation and had longer duration of hospital stay. We chose preterm infants < 32 weeks as these infants are more likely to be ventilated for a longer period of time, hence at risk of UE. Also, the outcome of interest is more relevant in this population. It was prudent to consider an appropriate matching cohort for the infants who experienced UE to ensure that both groups were exposed to similar risk factors associated with the outcomes of interest. For the primary analysis, we did not include 8 infants who had UE and did not require re-intubation. Most likely these infants were ready for extubation prior to the UE and were ventilated for longer than necessary. This population would not be truly representative of our study cohort who continued to need mechanical ventilation following the UE. The sensitivity analysis including all the infants who had UE showed similar results. These further confirms the strengths of our study results.

During the study period, a QI project in our center led to an 80% reduction in UE rates over an 18-month period (7.4 UE/100 ventilation days to 1.4 UE/100 ventilation days). The most common adverse events associated with UE were hypothermia (39%) and the need for cardiopulmonary resuscitation (6.8%).3 Higher rates of adverse events have been reported for emergent re-intubations following UE.7 These adverse events may lead to a need for escalation of cardiorespiratory support. This may also impact on the decisions around timing of future planned extubations that indirectly prolong mechanical ventilation.

Several studies have shown that prolonged duration of mechanical ventilation is associated with greater mortality and pulmonary morbidities like BPD, pulmonary hypertension, periventricular leukomalacia, ROP, and impaired neurodevelopment at 2 years.12-15 In our study, BPD at 36 weeks postmenstrual age was significantly higher in infants who experienced UE. After further adjustment for mechanical ventilation duration, UE was not associated with BPD. However, infants with longer duration of mechanical ventilation had higher odds of developing BPD as shown in other studies. We speculate that UE may indirectly increase the risk of BPD by prolonging the time spent on invasive ventilation.

Previous studies have shown the odds of ROP were higher with longer duration of mechanical ventilation.16,17 Similar to BPD, in a logistic regression model adjusting for mechanical ventilation duration, UE was not predictive of severe ROP in our study. But mechanical ventilation duration was significantly associated with development of severe ROP. The plausible reasons for this association may be that these infants have higher severity of illness,18 frequent fluctuations in oxygen levels while on the ventilator,19 and poor postnatal growth.20 We did not collect and analyze any of these variables in our study.

Similar to our study, Hatch et al6 also demonstrated that the greatest risk of UE was between 7–28 chronological age in days. In our study, 17 infants (36%) experienced UE in the first week of life. Two infants who experienced UE required surgical intervention for acquired subglottic stenosis (5%). Both these infants had multiple UEs. None of the infants in control group developed subglottic stenosis. Thomas et al showed that infants with severe, acquired subglottic stenosis were more likely to have had episodes of UE and more than 2 UE per 100 ventilation days.8 However, this was not statistically significant in a multivariate analysis in their study.

Our study has some limitations. The groups were not controlled for severity of illness (eg, clinical risk index for babies score) and severity of lung disease (eg, oxygenation index). However, variables used for matching, namely birth gestation, birthweight, and duration of mechanical ventilation prior to UE, are shown to have similar risk for the outcome of interest. The controls in the study were as closely matched as possible with cases. Despite this, some differences may still exist and could influence the outcome. Conducting a randomized trial on this topic would be unethical, and thus we considered using matched controls a suitable alternative. We did not perform any power analysis as we included all eligible cases in the study period associated with the QI project.

The strengths of the study include the data were prospectively collected for all ventilated infants through the QI project. We only included infants who had ventilation episodes exclusively in the study center. This avoided any center-associated variations in respiratory management. We excluded infants who were managed on noninvasive ventilation following UE as this may be indicative of the extubation readiness for these infants. The variables that were imbalanced after matching were used in the regression model.

There is a perceived notion that UE is associated with only immediate adverse outcomes. From our study, we have shown that UE is associated with major adverse hospital outcomes. Prevention of UE could be one of the effective interventions to reduce major morbidities.

Conclusions

In our study, UE was associated with significantly longer stay in the hospital and increased time spent on the ventilator. These may indirectly be associated with major morbidities like BPD and ROP. It is crucial to implement care bundles to reduce the rates of UE to help improve the outcomes in infants cared for in NICU.

Acknowledgments

We would like to thank the clinical audit team at South Tees Trust who helped us with data collection. We would like to thank Dr Francesco Cavallin at the University of Padua (Italy) for his statistical support and advice.

Footnotes

  • Correspondence: Vrinda Nair MD, Neonatal Intensive Care Unit, James Cook University Hospital, Marton Road, Middlesbrough, United Kingdom. E-mail: Vrinda.nair1{at}nhs.net

  • The authors have disclosed no conflicts of interest.

  • Supplementary material related to this paper is available at http://www.rcjournal.com.

REFERENCES

  1. 1.↵
    1. Veldman A,
    2. Trautschold T,
    3. Weiss K,
    4. Fischer D,
    5. Bauer K
    . Characteristics and outcome of unplanned extubation in ventilated preterm and term newborns on a neonatal intensive care unit. Paediatr Anaesth 2006;16(9):968-973.
    OpenUrlCrossRefPubMed
  2. 2.↵
    1. Merkel L,
    2. Beers K,
    3. Lewis MM,
    4. Stauffer J,
    5. Mujsce DJ,
    6. Kresch MJ
    . Reducing unplanned extubations in the NICU. Pediatrics 2014;133(5):e1367-e1372.
    OpenUrlAbstract/FREE Full Text
  3. 3.↵
    1. Nair V,
    2. Smith H
    . Phased quality improvement interventions in reducing unplanned extubation in the neonatal ICU. Respir Care 2020;65(10):1511-1518.
    OpenUrlAbstract/FREE Full Text
  4. 4.↵
    1. Kambestad KK,
    2. Huack A,
    3. Nair S,
    4. Chapman R,
    5. Chin S,
    6. Langga L,
    7. et al
    . The adverse impact of unplanned extubation in a cohort of critically ill neonates. Respir Care 2019;64(12):1500-1507.
    OpenUrlAbstract/FREE Full Text
  5. 5.↵
    1. Klugman D,
    2. Berger JT,
    3. Spaeder MC,
    4. Wright A,
    5. Pastor W,
    6. Stockwell DC
    . Acute harm: unplanned extubations and cardiopulmonary resuscitation in children and neonates. Intensive Care Med 2013;39(7):1333-1334.
    OpenUrlPubMed
  6. 6.↵
    1. Hatch LD,
    2. Grubb PH,
    3. Markham MH,
    4. Scott TA,
    5. Walsh WF,
    6. Slaughter JC,
    7. et al
    . Effect of anatomical and developmental factors on the risk of unplanned extubation in critically ill newborns. Am J Perinatol 2017;34(12):1234-1240.
    OpenUrl
  7. 7.↵
    1. Hatch LD,
    2. Grubb PH,
    3. Lea AS,
    4. Walsh WF,
    5. Markham MH,
    6. Whitney GM,
    7. et al
    . Endotracheal intubation in neonates: a prospective study of adverse safety events in 162 infants. J Pediatr 2016;168:62-66.e66.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Thomas RE,
    2. Rao SC,
    3. Minutillo C,
    4. Vijayasekaran S,
    5. Nathan EA
    . Severe acquired subglottic stenosis in neonatal intensive care graduates: a case-control study. Arch Dis Child Fetal Neonatal Ed 2018;103(4):F349-F354.
    OpenUrlAbstract/FREE Full Text
  9. 9.↵
    1. Roddy DJ,
    2. Spaeder MC,
    3. Pastor W,
    4. Stockwell DC,
    5. Klugman D
    . Unplanned extubations in children: impact on hospital cost and length of stay. Pediatr Crit Care Med 2015;16(6):572-575.
    OpenUrl
  10. 10.↵
    1. Hatch LD 3rd.,
    2. Scott TA,
    3. Slaughter JC,
    4. Xu M,
    5. Smith AH,
    6. Stark AR,
    7. et al
    . Outcomes, resource use, and financial costs of unplanned extubations in preterm infants. Pediatrics 2020;145(6).
  11. 11.↵
    1. Staffa SJ,
    2. Zurakowski D
    . Five steps to successfully implement and evaluate propensity score matching in clinical research studies. Anesth Analg 2018;127(4):1066-1073.
    OpenUrlPubMed
  12. 12.↵
    1. Jensen EA,
    2. DeMauro SB,
    3. Kornhauser M,
    4. Aghai ZH,
    5. Greenspan JS,
    6. Dysart KC
    . Effects of multiple ventilation courses and duration of mechanical ventilation on respiratory outcomes in extremely low birthweight infants. JAMA Pediatr 2015;169(11):1011-1017.
    OpenUrlPubMed
  13. 13.
    1. Choi YB,
    2. Lee J,
    3. Park J,
    4. Jun YH
    . Impact of prolonged mechanical ventilation in very low birthweight infants: results from a national cohort study. J Pediatr 2018;194:34-39.e33.
    OpenUrlPubMed
  14. 14.
    1. Vliegenthart RJS,
    2. van Kaam AH,
    3. Aarnoudse-Moens CSH,
    4. van Wassenaer AG,
    5. Onland W
    . Duration of mechanical ventilation and neurodevelopment in preterm infants. Arch Dis Child Fetal Neonatal Ed 2019;104(6):F631-F635.
    OpenUrlAbstract/FREE Full Text
  15. 15.↵
    1. Escobar V,
    2. Soares DS,
    3. Kreling J,
    4. Ferrari LSL,
    5. Felcar JM,
    6. Camillo CAM,
    7. et al
    . Influence of time under mechanical ventilation on bronchopulmonary dysplasia severity in extremely preterm infants: a pilot study. BMC Pediatr 2020;20(1):241.
    OpenUrl
  16. 16.↵
    1. Kim TI,
    2. Sohn J,
    3. Pi SY,
    4. Yoon YH
    . Postnatal risk factors of retinopathy of prematurity. Paediatr Perinat Epidemiol 2004;18(2):130-134.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Yang CY,
    2. Lien R,
    3. Yang PH,
    4. Chu SM,
    5. Hsu JF,
    6. Fu RH,
    7. et al
    . Analysis of incidence and risk factors of retinopathy of prematurity among very low birthweight infants in North Taiwan. Pediatr Neonatol 2011;52(6):321-326.
    OpenUrlPubMed
  18. 18.↵
    1. Gantz MG,
    2. Carlo WA,
    3. Finer NN,
    4. Rich W,
    5. Faix RG,
    6. Yoder BA,
    7. et al
    ; SUPPORT Study Group of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Achieved oxygen saturations and retinopathy of prematurity in extreme preterms. Arch Dis Child Fetal Neonatal Ed 2020;105(2):138-144.
    OpenUrlAbstract/FREE Full Text
  19. 19.↵
    1. Das A,
    2. Mhanna M,
    3. Sears J,
    4. Houdek JW,
    5. Kumar N,
    6. Gunzler D,
    7. et al
    . Effect of fluctuation of oxygenation and time spent in the target range on retinopathy of prematurity in extremely low birth weight infants. J Neonatal Perinatal Med 2018;11(3):257-263.
    OpenUrl
  20. 20.↵
    1. Hellström A,
    2. Ley D,
    3. Hansen-Pupp I,
    4. Niklasson A,
    5. Smith L,
    6. Löfqvist C,
    7. et al
    . New insights into the development of retinopathy of prematurity–importance of early weight gain. Acta Paediatr 2010;99(4):502-508.
    OpenUrlCrossRefPubMed
Back to top

In this issue

 

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section

Related Articles

Cited By...

Keywords