Abstract
BACKGROUND: Inhalation injury increases morbidity and mortality in burn patients. Patients with inhalation injury present with large differences between end-tidal CO2 pressure and 
, an indirect measure of dead space. We aimed to investigate the relationships between increased dead space and inhalation injury outcomes.
METHODS: This retrospective study included 51 adult subjects with burns and inhalation injuries. Demographics, size of burns, length of stay, ventilator days, blood gas results, end-tidal CO2 pressure, presence of ventilator-associated pneumonia, and mortality data were collected. Modified Baux scores and ratios of alveolar dead space to alveolar tidal volume (
/
) were calculated. Independent t tests were used to compare mean 
/
of survivors to that of subjects who died and between subjects with and without pneumonia. The relationships between 
/
and ventilator days or modified Baux score were assessed with bivariate correlation analysis.
RESULTS: Our population had a mean age of 52 y and an average burn size of 17.5%. The average length of stay and ventilator days were 12 d and 3.8 d, respectively. The mean modified Baux score was 87. The mean 
/
was 0.38. Ten subjects died, and 6 subjects had pneumonia. The 
/
of survivors was significantly smaller for survivors than for subjects who died (0.34 vs 0.52, P = .03). No significant difference was observed between subjects with and without pneumonia (0.36 vs 0.47, P = .26). 
/
correlated significantly with modified Baux score (r = .524, P < .001).
CONCLUSIONS: Alveolar dead space (
/
) is easily calculated from 
and end-tidal CO2 pressure and may be useful in assessing severity of inhalation injury, the patient’s prognosis, and the patient’s response to treatment.
- burn
- inhalation injury
- mortality
- ratio of dead space to tidal volume
- dead space fraction
- modified Baux scores
- total burn surface area
Introduction
Inhalation injury remains a significant cause of morbidity and mortality despite overall improvement in survival from burn injuries. The pathophysiology of inhalation injury involves direct, local injury to the tracheobronchial tree and humoral mediated derangements of the pulmonary microvasculature. The former results in respiratory epithelial sloughing, tracheobronchial hemorrhage, and airway occlusion. The latter results in inflammatory infiltrates, leaky capillaries, increased pulmonary lymph flow, and pulmonary edema.1-3
Early studies of inhalation injury detected impairments of pulmonary mechanics and ventilation-perfusion mismatch. In 1970, Garzon et al4 reported on five patients with smoke inhalation and detected reduced pulmonary compliance and increased work of breathing. Moylan et al5 used 133xenon scans to describe the ventilation-perfusion mismatch from air trapping secondary to small bronchiole obstruction in patients with inhalation injury.
Dead space (VD), however, received little attention in the field of research on inhalation injury. Pruitt et al6 made a passing comment about the elevated ratio of dead space to tidal volume (VD/VT) in patients with thermal injury and “progressive pulmonary insufficiency,” but they did not specifically attribute the observation to patients with inhalation injury. Trunkey7 repeated the observation without elaboration in 1978. Recently, Dries and Endorf8 commented on increased dead-space ventilation associated with inhalation injury in their review on the topic.
In our burn unit, the bedside monitors display and record continuous end-tidal CO2 pressure (
) for our patients on ventilators. These data, along with ventilator settings, ventilator measurements, and blood gas results, are uploaded to the respiratory therapy flowsheets and stored in the electronic medical record. These flowsheets provide detailed records of patients’ respiratory physiology. We noted abnormally wide differences between measured 
on arterial blood gas results and 
in some of our patients with inhalation injury. This difference between 
and 
can be used to calculate the alveolar dead space (
/
) as described by Hardman and Aitkenhead:9,10
where 
and 
are the alveolar VD and alveolar VT, respectively. The normal VDalv/VTalv in healthy adults is 0.15–0.35.
We undertook this study to estimate the prevalence of increased 
/
in patients with inhalation injury and its possible relationship with morbidity and mortality associated with inhalation injury.
Quick Look
Current Knowledge
Early studies of inhalation injury detected impairments of pulmonary mechanics and ventilation-perfusion mismatch. Dead space, however, received little attention in the field of research on inhalation injury.
What This Paper Contributes to Our Knowledge
Our results indicate that the ratio of alveolar dead space to alveolar tidal volume (VDalv/VTalv) was often increased in subjects with inhalation injury. In our population, VDalv/VTalv was higher in subjects who died and in those who suffered acute kidney injury, and VDalv/VTalv correlated moderately with total body surface area of burn and modified Baux Score.
Methods
Subjects
This retrospective study was approved by the University of Iowa Institutional Review Board. We queried the University of Iowa Hospital admission database for subjects with a diagnosis of inhalation injury using StarMaker software (Park Street Solutions, Naperville, Illinois). All patients admitted to our burn center and diagnosed with inhalation injuries from January 2000 to January 2016 were included in this study. Pediatric patients and patients who were misdiagnosed and did not have inhalation injuries were excluded from the study. Smoke inhalation injury was defined as evidence of epithelial damage below the vocal cords on bronchoscopy.
Data Collection
Data on the following characteristics were abstracted from the medical records of adult subjects diagnosed with inhalation injuries: age, sex, height, weight, body mass index, total body surface area, total body surface area of burn, date and time of burn, date and time of admission, history of lung disease, smoking status, carboxyhemoglobin, presence of ventilator-associated pneumonia (VAP) or acute kidney injury during the hospital stay, arterial blood gas results, 
, ventilator settings and records, date and time of liberation from the ventilator, and vital signs.
We used the collected data to derive the following variables: Baux score (age + total body surface area of burn), modified Baux score (age + total body surface area of burn + 17),11 CO2 gap (
– 
), and 
/
.9,10 The CO2 gap was calculated using arterial blood gas results from a sample that had been drawn within 20 min of a recorded 
reading. Many subjects had multiple gap results because of their long period of ventilator support. Thus, we considered all 
and 
available in the medical record. 
was captured and stored by the monitor at a frequency of 30–120 min. 
from arterial blood gas was recorded on admission and as needed depending on the attending physician preference or based on the patient condition. The time at which the arterial blood gas specimen was obtained was recorded for the results in the medical record. We reasoned that the result of the arterial blood gas would not be available to the physician or respiratory therapist until 10–20 min after the blood draw to account for transportation to the lab, completion of the lab analysis, and reporting the results. Therefore, even if ventilator changes were made based on the lab results, we thought the 
was an accurate estimate for at least 20 min from the time of the blood draw. 
/
was calculated using the maximum difference between 
and 
. Ventilator days were calculated using the date and time of ventilator liberation.
Statistical Analysis
Descriptive statistics were obtained. Independent t tests and nonparametric Mann-Whitney tests were used as appropriate. Bivariate correlation was performed to assess the association between 
/
and other variables. Multivariate linear regression analysis was performed to assess the impact of smoking and chronic lung disease on the association between 
/
and other variables. Statistical analysis was performed using SPSS 25.0 (IBM SPSS, Armonk, New York). Significance was considered at P < .05.
Results
Subject Characteristics
A total of 57 patients diagnosed with inhalation injuries were identified. After review of their medical records, we excluded 6 patients: 3 were pediatric patients, 1 patient had 2 medical record numbers and required merging the data into a single case, and 2 patients who were misdiagnosed and did not have inhalation injuries.
Descriptive statistics for the population are presented in Table 1. The population was composed of 31 men (61%); 24 subjects were known cigarette smokers (47.1%), and 12 had a history of COPD (23.5%). The mean subject age was 52.3 y. The average burn size was 17.5%, and the average modified Baux Score was 86.8. The mean length of stay and mean ventilator days were 11.9 d and 3.8 d, respectively. The mean carboxyhemoglobin was 6.4 among the 34 subjects for whom it was measured. 
/
ranged from 0.05 to 0.97; the average was 0.38, slightly above the normal range. On average, 
/
was not significantly different between smokers and nonsmokers (0.46 ± 0.21 vs 0.34 ± 0.25, respectively; P = .39), subjects presenting with chronic lung diseases (including COPD) or not (0.40 ± 0.24 vs 0.36 ± 0.23, respectively; P = .75), or subjects who developed ARDS (n = 9; 17.6%) assessed on the basis of the Berlin Criteria12 or not (0.37 ± 0.23 vs 0.39 ± 0.20, P = .70). ARDS was determined using the Berlin Criteria because a definitive clinical diagnosis of ARDS was noted in the medical records of only 5 subjects.
Subject Characteristics
Ten subjects died (19.6%), and 41 survived (80.4%). Six subjects suffered VAP (11.8%), but they all survived. Thirteen subjects developed acute kidney injury (25.5%); 6 died, and 7 survived.
/ and Outcomes
For our analysis on VDalv/VTalv and outcomes, only 41 subjects had appropriate 
and 
measurements to make the calculations. In the other 10 cases, either 
or 
was not available. As shown in Table 2, the mean 
/
was significantly lower in subjects who survived compared to that of subjects who died. Similarly, the mean 
/
was significantly higher in subjects who suffered acute kidney injury than in those who did not have acute kidney injury. No significant difference was observed in 
/
between subjects with and without VAP.
/
and Outcomes
Length of stay and ventilator days correlated weakly, but not significantly, with 
/
(r = 0.276 and 0.281, respectively; P = .08). No correlation was observed between 
/
and body mass index, 
, or the 
/
ratio. As shown in Figure 1, both total body surface area of burn size and modified Baux score correlated moderately but significantly with 
/
(r = 0.606 and 0.524, respectively; P < .001). Multivariate regression analysis indicated that being a smoker, presenting with a history of chronic lung disease or presenting with ARDS did not affect the association between burn size and modified Baux scores with 
/
(data not shown).
Ratio of alveolar dead space to alveolar tidal volume (
/
) correlates with (A) total body surface area of burn and (B) modified Baux score.
Discussion
This was a retrospective review of subjects with inhalation injury from a single institution. Detailed physiologic data were available from the electronic medical records, particularly continuous 
measurements and laboratory results. These data allowed for the easy calculation of 
/
. 
/
is another measure of respiratory physiology and is often increased in patients with inhalation injury. We calculated the alveolar dead space because the equation uses the 
rather than the mixed expiratory CO2 that is used in the Enghoff modification of the Bohr equation. The latter requires extra apparatus on the ventilator and cumbersome calculations. In this study, subjects who died and subjects who suffered acute kidney injury presented with significantly increased 
/
values.
Early studies on inhalation injury noted the impaired ventilation and marked ventilation-perfusion mismatch in many of these patients.5,7 Pruitt et al6 mentioned increased VD/VT in burn subjects with inhalation injury and “progressive pulmonary insufficiency,” but they did not elaborate on the finding. They used the terms “progressive pulmonary insufficiency” instead of “shock lung” from the Vietnam era. Their description of progressive pulmonary insufficiency resembles ARDS, which was not a commonly used term at that time.
Further research into inhalation injury elaborated on the role of large-volume fluid resuscitation and systemic inflammatory mediators in the pathophysiology of inhalation injury. Peitzman et al13 reported increased extravascular lung water in subjects with inhalation injury who died of sepsis. In 1996, Darling et al14 implicated large-volume fluid resuscitation in worsening pulmonary compliance in subjects with inhalation injury.
The role of systemic inflammatory mediators in secondary pulmonary injury in burns was identified by Rue et al15 in 1993. They discovered that inflammatory mediators from cutaneous burns circulate via the bloodstream and damage the lungs. Secondary lung injury in inhalation injury was concisely summarized in reviews by Cancio and colleagues in 20051 and 2007.16 In the latter review, the authors confirmed that the ventilation-perfusion mismatch in inhalation injury is not a shunt.1
In more recent studies, investigators identified increased VD/VT in subjects with ARDS. Nuckton et al17 used the Enghoff modification of the Bohr equation to calculate the VD/VT in 179 subjects and reported that VD/VT predicted mortality. In 2014, Kallet et al18 studied VD/VT in subjects with acute lung injury and ARDS and found that VD/VT rose early during the disease and was associated with increased mortality.
These studies suggest that increased VD/VT is associated with a variety of pulmonary disorders and is not unique to inhalation injury. Possible mechanisms for the increased VD include hyperexpansion with reduced alveolar capillary flow or reduced pulmonary vascular flow secondary to endothelial damage.19 Another plausible mechanism is incomplete expiration due to “dynamic hyperexpansion” with auto-PEEP as de-scribed by Rees and colleagues.20
In this study, increased VD/VT was associated with mortality and acute kidney injury, but not with ARDS, which is a potential confounding factor. These results suggest that the systemic effects of burn injuries on inhalation injury play a role in increasing VD/VT. One could propose that larger burn sizes generate more inflammatory mediators and require larger volumes of fluid resuscitation. The former might cause more distal organ injury, and the latter would increase extravascular lung water. Both mechanisms have been implicated in lung injury and mortality in patients with inhalation injury. In our study, the moderate correlation between 
/
and modified Baux Score, in which burn size is a significant factor, may be caused by either or both of these mechanisms. The stronger correlation between burn size and 
/
might discount age as a factor and favor the magnitude of the burn injury. Additionally, increased 
/
did not correlate with 
, suggesting that the increase in 
/
is more likely associated with impaired ventilation and not oxygenation. This study, however, does not provide sufficient evidence to reach that conclusion.
The stronger association between 
/
and acute kidney injury compared to the association between 
/
and VAP suggests systemic rather than local pulmonary mechanisms.
Our study has several limitations. First, the data were acquired with a retrospective chart review, which introduces a risk of bias due to poor documentation or missing data. Second, our sample size was small. Finally, only 6 subjects had VAP, and all of them survived. This group likely skewed the results in favor of survivors. Despite these limitations, our study suggests patterns of complex pathology in subjects with burns and inhalation injury, although the conclusions we can draw from our results are limited.
Conclusions
This study supports the feasibility and ease of calculating the alveolar VD/VT from continuous 
and 
measurements, both readily available at the bedside in a burn unit with modern monitors. Although we do not know how using 
/
may improve patient care, previous studies suggest that extravascular lung water may play a role, thus, diuresis or high PEEP may be beneficial.15 Conversely, lung-protective strategies and reduced PEEP may reduce VD and improve patient outcomes. Additionally, in our population, 
/
was significantly higher in subjects who died and in subjects with acute kidney injury. Finally, 
/
correlated with total body surface area of burn and modified Baux score. Considering all that we learned from this study, we suggest that calculation of 
/
might be useful to assess inhalation injury severity, outcomes, and response to treatment. This study paves the way for future prospective studies assessing the association between 
/
and outcomes.
Footnotes
- Correspondence: Thomas Granchi MD MBA, Department of Surgery, Acute Care Surgery Division, University of Iowa Hospitals and Clinics, 200 Hawkins Dr, Iowa City, IA 52242. E-mail: tgranch{at}emory.edu
At the time of this study, all authors were affiliated with the Department of Surgery, Acute Care Surgery Division, University of Iowa Hospitals & Clinic Iowa City, Iowa.
The authors have disclosed no conflicts of interest.
- Copyright © 2020 by Daedalus Enterprises
