High incidence of adult respiratory distress syndrome associated with amphetamine use in the burn population: a retrospective cohort study

Background:

A high incidence of adult respiratory distress syndrome (ARDS) is observed in patients with a history of amphetamine use with limited studies addressing this exact topic. The authors sought to understand and compare the clinical features of patients experiencing amphetamine-associated lung injury with features of similar patients naïve to amphetamines, in a population of burn patients. Patients in this population tend to be young with few comorbidities offering a unique opportunity to study the relationship between amphetamine use and ARDS.

Materials and methods:

A total of 188 patients age 18 and older with total body surface area (TBSA) between 20 and 60% were sampled over 5 years. To capture the moderate to severe burn population, a lower limit of 20% was chosen while 60% was used as the upper limit to exclude patients likely to die from the burns alone. Patients eligible to be include in the study had to meet the TBSA criteria. Demographic data was ascertained. Patients were placed into two cohorts: the amphetamine positive group (AmPOS), the amphetamine negative group (AmNEG). Key endpoints included hospital mortality, length of ICU stay, development of ARDS, and cardiac output parameters. Nonparametric data was evaluated with the Mann–Whitney test and categorical variables were compared using χ ².

Results:

Data from 49 patients with ARDS were collected retrospectively out of the 188 patients in this TBSA range. The incidence of amphetamine abuse in these burn patients is 14.9%. The average age of AmPOS and AmNEG patients was 36 and 34 and the average TBSA of burns in the AmPOS and AmNEG group was 51.8 and 45.2%. The mean onset of ARDS was 2.2 days for the AmPOS versus 3.3 days for the AmNEG (P=0.19). At admission, patients with amphetamine use exhibited less inhalational injury and a lower Acute Physiology and Chronic Health Evaluation II (APACHE II) score. In AmPOS, 64% developed ARDS compared to 19% in AmNEG (P<0.001). Mortality, time on a ventilator, ICU days, packed Red Blood Cell, Fresh Frozen Plasm, platelets transfused, and initial cardiac parameters all were not statistically significant. On the initial date of ARDS diagnosis, there was no statistically significant difference with PaO₂/FiO₂ and compliance (P=0.67), but positive end expiratory pressure requirements was higher in AmPOS (P=0.018).

Conclusions:

Amphetamine Use was associated with increased risk of developing ARDS in the burn population. This is despite having a better APACHE II and lower incidence of inhalational injury in the AmPOS group, supporting amphetamine as an independent risk factor for ARDS.

Introduction

Highlights

• Amphetamine use significantly increases the risk of developing acute respiratory distress syndrome in the burn population.

• The occurrence of acute respiratory distress syndrome in the amphetamine positive population was 64% compared to 19% in the amphetamine negative population. This was out of proportion to the total burn surface area, inhalational injury component, and APACHE II score and is independent of the cardiac and volume effects.

Amphetamine affects both the cardiovascular and central nervous systems directly through alpha-receptors and beta-receptors and indirectly through the accumulation of the neurotransmitter dopamine and release of norepinephrine stores1. Amphetamine-related drugs are typically abused for their psychostimulatory characteristics. Currently, an estimated 4 million persons in the United States have abused drugs related to amphetamine. From 1991 to 1994, the number of methamphetamine-related deaths reported by medical examiners nearly tripled2. These deaths are usually caused by cardiovascular collapse. While amphetamine abuse has been noted to be a potential cause of lung injury, reports are anecdotal or case reports3.

A high incidence of acute respiratory distress syndrome (ARDS) in our surgical and medical patients with a history of amphetamine use has been observed. Geographically, Southern California is one of the largest regions with methamphetamine exposure4. Neeki et al. 5,6 showed that 40.7% of burn injuries in a 10-year time frame tested positive for amphetamine. These patients tend to be young with little comorbidity, offering a unique opportunity to study the relationship between amphetamine use and ARDS with few confounding factors such as sepsis, multiple medical comorbidities, and multiorgan dysfunction. The goal of this study is to further investigate the role amphetamine has in the development of ARDS and the clinical outcomes associated with it. We hypothesize that amphetamine use is associated with a higher incidence of ARDS in this burn population and this may provide clinicians further insight in applying future approaches to the management of these cases.

Materials and methods

This study was completed at a County Burn Center and American College of Surgeons-verified level I trauma center. Due to the high prevalence of methamphetamine use, this location sees a high incidence of burn patients who either use amphetamine or are the result of methamphetamine lab explosions5,6. The burn database and electronic patient record system captures all admissions to the health system including patient demographics, presentation, morbidity, and mortality. Consent, ethics approval, data gathering, and analysis was obtained from the Institutional Review Board (IRB #15-24) and collected in a password protected Excel document. No patient data was identified in this article. The work was reported in line with the STROCSS criteria7. This project has also been registered with clinicaltrials.gov/show/NCT05611385.

A complete retrospective chart review was performed for all the patients’ records and included all patients aged 18 years or higher who presented with total body surface area (TBSA) burns between 20 and 60% over 5 consecutive years. The TBSA range was chosen to identify patients with significant burns requiring formal fluid resuscitation and patients with TBSA greater than 60% were excluded due to a known higher mortality rate. In this study, patients with ARDS were diagnosed with a ratio of partial pressure of arterial oxygen to fractionally inspired oxygen (PaO₂/FiO₂) less than 300 and radiological finding of bilateral pulmonary infiltrate that occurred within five days of admission. The data collected from the first 5 days were chosen to exclude secondary causes of ARDS. Demographics, amphetamine use, Acute Physiology and Chronic Health Evaluation II (APACHE II), inhalational component, mortality and morbidity, respiratory, and cardiovascular data were retrospectively collected. Inhalation injury is defined as burn to the face including singed hair, oral, or nasal soot. Cardiovascular parameters cardiac output, pulmonary wedge pressure, pulmonary diastolic pressure, and central venous pressure were collected from pulmonary artery catheters. Pediatric patients, as well as patients with a history of vaping were excluded.

Patients were considered to use amphetamine if the urine drug screen (Roche Diagnostics) tested positive. The sample was subsequently divided into two cohorts: participants that were amphetamine positive and amphetamine negative. Patients only positive for amphetamines were included in the study. Ventilator data was used to evaluate the intrinsic pulmonary mechanical differences between patients with and without amphetamine abuse. The samples were furthered stratified to only consider the patients with ARDS to better sample our cohort of interest. Additional parameters considered in this analysis included demographics, amphetamine use, inhalation injury component, APACHE II and its associated chronic comorbidity parameters, mortality, respiratory, and cardiovascular data. Nonparametric data were evaluated with the Mann–Whitney test and categorical variables were compared using χ ². Statistical significance was determined by P less than 0.05. In calculating the P value, equal variance is not assumed. Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS; SPSS Inc. IBM).

Results

In this analysis, 188 patients were found to have a TBSA between 20 and 60%. The average age in the amphetamine positive (AmPOS) group was 36±2 compared to 34±2 in the amphetamine negative (AmNEG) group (P=0.58). Seventy-nine percent were male in the AmPOS group and 65% in the AmNEG group (P=0.54). Of this population, 28 patients or 14.9% tested positive for amphetamine. There were 64% or 18 patients with ARDS in the AmPOS group compared to 19% or 31 patients in the AmNEG, as shown in Figure 1. The occurrence of ARDS in the AmPOS group was statically significantly higher than the AmNEG group (P<0.001). Demographics and mortality between the AmPOS and AmNEG groups with and without ARDS are listed in Table 1. Note that there was a statistically smaller percentage of patients with inhalation injury in the AmPOS compared to the AmNEG group [6 (33%) vs. 21 (67%) P=0.03]. Of the patients that died, the three main documented causes of death were multisystem organ dysfunction, sepsis, and ARDS at 40, 11.4, and 25.7%, respectively.

Figure 1.

Flow diagram detailing the patient selection in the amphetamine positive and negative group. *Incidence of adult respiratory distress syndrome (ARDS) was statistically higher in the amphetamine tested positive group compared to the tested negative group (P<0.001).

Table 1.

Clinical demographics of amphetamine positive and amphetamine negative patients with and without ARDS

	Amphetamine positive (n=28)	Amphetamine negative (n=160)	P value
Age, years, averagea	36±2	34±2	0.58
Male [n (%)]	22 (79)	87 (65)	0.54
ARDS [n (%)]	18 (64)	31 (19)	0.001
Onset of ARDS, days, average	2.2±0.4	3.3±0.7	0.19
ARDS patients	(n=18)	(n=31)
PaO2/FiO2 a	112.9±19.3	126.0±24.1	0.67
ICU days, averagea	16.7±4.0	26.5±4.4	0.11
Ventilator days, averagea	13.4±2.6	21.3±3.4	0.07
Mortality [n (%)]	15 (83)	19 (61)	0.14
Total burn surface areaa (%)	51.8±4.6	45.2±3.3	0.26
Inhalational component [n (%)]	6 (33)	21 (67)	0.03
APACHE II a	10.9±1.4	14.7±1.1	0.04
Red blood cells, unitsa	15.6±6.2	16.3±2.5	0.93
Fresh frozen plasma, unitsa	10.5±3.6	7.8±2.1	0.52

^a Analyzed with Mann–Whitney test. All other comparisons with χ ²-test.

Respiratory parameters were also collected. The mean onset of ARDS in the AmPOS versus AmNEG was 2.2±0.4 versus 3.3±0.7 days (P=0.19), respectively. No statistical difference was found between the groups regarding PaO₂/FiO₂ (P=0.67) and ICU time in days (P=0.11). The average PaO₂/FiO₂ for AmPOS was 112.9±19.3 and it was 126.0±24.1 for AmNEG (P=0.67). There was a noted statically higher positive end expiratory pressure in the AmPOS group (9.6±1.2 vs. 7.4±1.0 mm Hg; P=0.02).

For the cardiovascular parameters, 12 of 28 AmPOS patients had pulmonary artery catheters with ARDS and 9 out of 31 patients in AmNEG. The cardiac output, wedge pulmonary diastolic pressure, and central venous pressure are listed in Table 2 and this study reveals no statistical significance between the two groups. None of the patients in either group had elevated cardiac enzymes suggesting myocardial injury.

Table 2.

Cardiovascular and respiratory parameters for amphetamine positive and amphetamine negative patients

	Amphetamine positive (n=12)	Amphetamine negative (n=9)	P value
Cardiac output (l/min)	7.0±1.1	10.3±1.3	0.08
Pulmonary wedge pressure (mm Hg)	19.4±2.5	16.9±1.3	0.39
Pulmonary diastolic pressure (mm Hg)	32.0±4.6	26.8±2.7	0.38
Central venous pressure (mm Hg)	19.5±2.5	15.8±1.6	0.23
Positive end expiratory pressure (mm Hg)	9.6±1.2	7.4±1.0	0.18

Analyzed with Mann–Whitney test.

Discussion

The Counties of San Bernardino, Inyo, Mono, and Riverside provide a relatively high number of amphetamine users per capita5. The average age of 36 at the time of burn injury in this data signifies a low degree of comorbidities and risk factors for individuals developing ARDS, allowing isolation of amphetamine use and ARDS interaction. Most literature reports on the various cardiovascular effects of amphetamine8, but this study reviews an association with lung injury and amphetamine use. A lower APACHE II score and lower incidence of inhalational injury should correlate with lower ARDS. However, the AmPOS group had a significantly higher incidence of ARDS, at 64.3 vs. 19.4% (P=0.00). This supports the notion that amphetamine is possibly an important contributor to the development of ARDS independent of APACHE II score. While mortality was not significantly higher with AmPOS group, it is most likely due to the higher APACHE II score and higher incidence of inhalation injury in the AmNEG group predicting a higher mortality. Inhalation injury is an independent predictor of mortality based on the Baux score for burn mortality. The longer ventilator days in the AmNEG group is consistent with the higher incidence of inhalation injury in this group. This suggests that if both groups had similar risk factors, the AmPOS patients may have a higher mortality.

ARDS, as defined by this study, is consistent with the Berlin Definition but did not include the 1994 European-American Consensus Conference definition of pulmonary wedge pressure less than 18 mm Hg because pulmonary catheters were not used on everyone9. However, analysis of the patients with a pulmonary catheter in place between the groups was not consistent with wedge pressures over 18 mm Hg. Amphetamine use has known mechanisms for cardiac failure, as severe arterial vasoconstriction resulting from acute amphetamine intoxication can lead to end organ ischemia with acute myocardial infarction and death10,11. Previous postmortem forensic studies have shown that methamphetamines have higher concentration in the left ventricle than the right ventricle, with higher concentrations possibly leading to cardiogenic adverse effects12. The possibility of myocardial necrosis and failure resulting in cardiogenic pulmonary edema occur from the imbalance of the cellular redox equilibrium with a surplus of reducing substances increased from amphetamine intoxication13. While it is possible the amphetamine effects on the cardiovascular system could lead to cardiogenic bilateral infiltrates, this study did not support the idea that bilateral pulmonary infiltrates were cardiac in nature, as there was no statistically significant difference between the cardiovascular parameters. The corrected pulmonary wedge pressure was less than 16 mm Hg, suggesting noncardiogenic origin. We also did not find a higher incidence of myocardial injury with cardiac enzymes.

Large volume fluid resuscitation can lead to pulmonary edema and may bias this study. Colloid use, wedge pressure, diastolic pulmonary pressure, and central venous pressures between the two groups were not statistically different, arguing against volume as a potential bias. Fluid resuscitation was under protocol based on Modified Parkland’s Formula including the use of colloid after initial crystalloid resuscitation and was constant across both groups. We felt that the wedge pressure, diastolic pulmonary pressure, central venous pressure, and positive end expiratory pressure provided a good surrogate to volume resuscitation, since measuring actual volume given may not be reflective of the intravascular needs in this burn patient population. Furthermore, these parameters were not statistically significant and thus indicated that the incidence of ARDS reported in this study is independent of volume or cardiogenic effects. It should be noted that pulmonary catheter comparisons did not include all patients. Pulmonary catheters were only used if the patient did not respond clinically to volume resuscitation. A type II statistical error is possible with the small sample size where a difference was not detected when there truly was a difference. However, the colloid resuscitation requirements (fresh frozen plasma) measured in all the patients did not show a difference (P=0.52), which supports the same conclusion from the pulmonary artery catheter comparison.

The plausibility behind amphetamine induced ARDS could be described by the direct effects to the lung as possible mechanisms of injury. Physiologic studies of methamphetamines and its derivatives have structural relationships that have an affinity to lung basement membrane14. Amphetamine inhibits potassium channels on the vascular smooth muscle of the pulmonary artery causing vasospasm with resulting pulmonary hypertension15. Another route of injury to the lung may be secondary to increased inflammatory response. Nuclear factor Kappa-B is a regulator of transcription factors associated with numerous inflammatory cascades. Amphetamine has been shown to induce nuclear factor Kappa-B in vitro 16. In addition, chronic use of amphetamine may downregulate receptors resulting in hemodynamic shock, which can contribute to the inflammatory response and therefore increase the risk for ARDS17.

There were several limitations to this retrospective study. Inhalation injury was subjectively diagnosed in the charts reviewed. Common data points used to define inhalational injury was not consistently documented. Instead, facial burns were chosen as a surrogate for inhalation injury because it was objective and consistently documented in the Lund–Browder charting required through institution protocol. Additionally, not all comorbidities were included. We included comorbidities that was required in the calculation of the APACHE II score which includes cirrhosis, New Your Heart Association Class IV, severe pulmonary disease, dialysis, and immunocompromised states. We felt these comorbidities in addition to burn parameters like TBSA and inhalation injury was the more accurate method to stratify the two populations in a retrospective study. Other comorbidities may play a role like diabetes and smoking history but is beyond the scope of this study with a small sample size.

Another limitation involved determining chronicity of amphetamine use. Whether the observed lung effects are due to acute or chronic ingestion is difficult to assess since the patients usually hide the abuse. The urine drug screen is also not quantitative, making it difficult to measure dose effect. Depending on the route of intake, the effects of amphetamine typically last less than 10 h18. In fact, ARDS could even be due to the withdrawal of amphetamine, and not from the acute intoxication since most of these patients developed ARDS beyond 10 h from the last drug use. Future studies should clarify this possible mechanism and may also need to look at drugs that can treat both the acute intoxication and withdrawal from amphetamine use.

Conclusion

The role amphetamine has in the development of ARDS and the clinical outcomes associated is a complex mechanism. Our young burn population and high incidence of amphetamine abuse allowed us to isolate amphetamine as a primary contributor to the development of ARDS. There was a noted association with the occurrence of ARDS in the AmPOS group being significantly larger and independent of TBSA burn, inhalational injury component, and APACHE II score. Additionally, a strong relationship between amphetamine use and ARDS appears to exists that is independent of the cardiac and volume effects.

Ethical approval

Ethics approval granted from Arrowhead Regional Medical Center Institutional Review Board (IRB #15-24).

Consent

All data anonymized and confidential.

Sources of funding

The research presented in this manuscript had no specific funding from any agency in the public, commercial or not-for-profit sectors.

Author contributions

D.T.W. contributed to study design, data collection, management of the patients, literature review as well as writing, and editing of the manuscript. A.M. contributed to study design, manuscript write-up, literature review, and editing of the manuscript.

Conflicts of interest disclosure

The authors declare there is no conflict of interest.

Research registration unique identifying number (UIN)

1. Name of the registry: Clinicaltrails.Gov

2. Unique identifying number or registration ID: NCT05611385

3. Hyperlink to your specific registration: https://clinicaltrials.gov/show/NCT05611385

Guarantor

David T. Wong, MD.

Provenance and peer review

Not commissioned, externally peer-reviewed.