Factors associated with mortality in acute respiratory failure patients without acute respiratory distress syndrome

Introduction

Mechanical ventilation (MV) is an indispensable treatment for acute respiratory failure patients, although its pulmonary and hemodynamic complications are well-known (1). Standard recommendations for MV strategy are available for patients with acute respiratory distress syndrome (ARDS) (2). Lung-protective ventilation strategies using low tidal volume, 4–8 mL/kg predicted body weight (PBW), and limited plateau pressure have reduced mortality among ARDS patients (2,3). For those with moderate to severe ARDS, prone positioning, referring patients who meet the criteria for extracorporeal membrane oxygenation (ECMO) to ECMO centers, and avoiding high-pressure recruitment maneuvers are recommended to reduce mortality (2-5). Recent studies also revealed associations between high driving pressure and mechanical power and increased mortality in ARDS patients (6-8). However, only 6–8% of patients requiring MV had ARDS at the start of MV, and ARDS accounts for only 23% of mechanically ventilated patients in the intensive care unit (ICU) (2,9). Most patients receiving invasive MV do not have ARDS; therefore, data on appropriate ventilation strategies in these patients are essential. This cohort study aimed to determine factors associated with increased mortality among acute respiratory failure patients without ARDS. We focused on the MV parameters that can be adjusted to improve patient outcomes. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-24-58/rc).

Methods

Study population

We prospectively collected the data of all consecutive patients aged ≥18 years who did not meet the diagnostic criteria for ARDS according to the Berlin definition (10) and were admitted to the medical ICU and received MV at Siriraj Hospital, Bangkok, Thailand, from October 2017 to September 2018. Patients intubated more than 24 hours before transfer to the ICU or those who had tracheostomy before admission were excluded.

This study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study was approved by the Human Research Protection Unit of the Faculty of Medicine Siriraj Hospital, Mahidol University (Certificate of Approval No. Si 546/2017). Written informed consent was obtained from all patients or their legal guardians if the participants could not provide the consent.

Data collection

The collected data included patient demographics, principal diagnosis of acute respiratory failure, comorbid diseases including diabetes mellitus, hypertension, chronic obstructive pulmonary diseases and chronic lung diseases, asthma, chronic kidney disease, cirrhosis, malignancy, coronary artery disease, cerebrovascular disease, immunosuppression, and smoking, sequential organ failure assessment (SOFA) score (11), baseline dynamic lung compliance (C_dyn) and lung injury score (LIS) (12), arterial blood gas and initial MV parameters at ICU admission, arterial blood gas and the ratio of partial pressure of oxygen in arterial blood to the fraction of inspiratory oxygen concentration (PaO₂/FiO₂) during invasive MV on the first day of ICU admission, ICU treatments during the first 3 days of ICU admission including MV parameters, renal replacement therapy and the use of fentanyl, midazolam, propofol, and neuromuscular blocking agent. The collected MV parameters during the first three days of ICU admission included ventilator mode, tidal volume (TV) in mL/kg PBW, peak inspiratory pressure (PIP), positive end-expiratory pressure (PEEP), dynamic driving pressure (ΔP) defined as inspiratory pressure set in pressure-controlled ventilation mode or the difference between PIP and PEEP in volume-controlled ventilation mode, the set and actual respiratory rates, minute ventilation, and fraction of inspired oxygen (FiO₂). Dynamic respiratory system compliance (C_dyn) is a ratio of tidal volume and the difference between PIP and PEEP. We recorded the parameters for the most prolonged duration of each day for all MV variables. We averaged variables during the first three days of ICU admission to represent ventilator strategies provided to the patients.

Outcomes

The primary outcome was all-cause 28-day mortality. Secondary outcomes were respiratory complications, including arterial pH, PaO₂/FiO₂, PaCO₂, and HCO₃, and C_dyn on day 3 of ICU admission, LIS on day 7, worsening lung injury defined as an increase in LIS of ≥0.25 points from baseline LIS at day 7 of ICU admission, newly developed pneumothorax, ventilator-associated pneumonia or ARDS, and use of rescue therapy for refractory hypoxemia, including prone positioning, recruitment maneuvers, or ECMO.

Statistical analysis

To identify factors associated with increased mortality in acute respiratory failure patients without ARDS, we compared data between survived and deceased patients at 28 days of admission. Continuous data were presented as mean ± standard deviation, median [interquartile range (IQR)] as appropriate, while categorical data were described as counts and column percentages. Univariate two-group comparisons were performed using independent two-group t-tests for normally distributed continuous variables, Mann-Whitney U tests for non-normally distributed continuous variables, and Chi-square or Fisher exact tests for categorical variables. Multivariable logistic regression modeling was performed to identify independent risk factors for 28-day mortality, and adjusted odds ratios (ORs) with 95% confidence intervals (CI) were reported. We planned to include all baseline characteristics and treatment variables reaching statistical significance in univariate analysis in the multivariable model. Statistical significance is set at P<0.05. All analyses were performed with SPSS version 18 (13).

Results

Study population

From October 2017 to September 2018, 342 patients were admitted to the medical ICU at Siriraj Hospital; 289 patients required MV and did not meet the diagnostic criteria for ARDS according to the Berlin definition. One hundred fifty-five patients were excluded, 154 patients were intubated ≥24 hours before transferring to ICU, and one had a tracheostomy placed before admission. One hundred and thirty-four patients were included in the final analysis; 69 patients (51%) died within 28 days, and 65 (49%) survived (Figure 1).

Figure 1 Patient population. ARDS, acute respiratory distress syndrome; ICU, intensive care unit.

Baseline characteristics

The median age of patients was 64 years, and 57% were male. Septic shock (59%) is the most common principal diagnosis, followed by pneumonia (37%) and heart failure (5%), with no significant difference in the principal diagnosis between survived and dead patients (Table 1). Compared with patients who survived, patients who died were more likely to have underlying cerebrovascular diseases (20% vs. 5%, P=0.006) and had higher baseline SOFA scores [11 (IQR, 8–14) vs. 8 (IQR, 6–12), P<0.001]. Other patient demographics were not significantly different between the two groups (Table 1).

For the baseline respiratory variables, 115 patients (86%) received pressure-controlled ventilation, and 19 patients (14%) received volume-controlled ventilation at ICU admission with a median TV and PIP of 8.2 (IQR, 6.8–10) mL/kg PBW and 23 (IQR, 19–26) cmH₂O, respectively. Deceased patients had lower baseline C_dyn [26 (IQR, 20–32) vs. 30 (IQR, 23–40) mL/cmH₂O, P=0.02] and were ventilated with higher ΔP [18 (IQR, 14–20) vs. 16 (IQR, 14–20), P=0.03] compared with survived patients. Other respiratory variables, arterial blood gas results, and LIS on admission were not significantly different between alive and dead patients; the median PaO₂/FiO₂ ratio was 258 (IQR, 161–413). The MV settings and outputs and other respiratory variables on the first day of ICU admission are shown in Table 1.

Treatments during the first three days of ICU admission

Patients who died within 28 days were ventilated with higher average PIP [24 (IQR, 21–27) vs. 21 (IQR, 19–25), P=0.001], ΔP [17 (IQR, 15–21) vs. 15 (IQR, 13–18), P<0.001], and actual respiratory rate [22 (IQR, 19–25) vs. 19 (IQR, 17–23), P=0.008] during the first 3 days of ICU admission, compared with those who were alive. Other MV parameters during the first three days were not different between the two groups, with a median TV of 8.1 mL/kg PBW, PEEP 5 cmH₂O, and FiO₂ 0.4. There were no significant differences in the administration of sedative and neuromuscular blocking drugs and receipt of renal replacement therapy in both groups (Table 2).

Respiratory parameters and complications

Compared with survived patients, deceased patients had higher LIS [1.5 (IQR, 1–2.25) vs. 0.67 (IQR, 0–1.25), P<0.001] and were more likely to have worsening LIS (43% vs. 17%, P=0.009) on day 7. There were no significant differences in the incidence of pneumothorax, ventilator-associated pneumonia, new-onset ARDS, and rescue therapy for refractory hypoxemia (Table 3).

Factors associated with hospital mortality in multivariable analysis

In multivariable analysis, underlying cerebrovascular diseases (OR 7.09, 95% CI: 1.78–28.28, P=0.006), higher SOFA score (OR 1.15, 95% CI: 1.04–1.28, P=0.008), and higher average PIP used during the first three days of admission (OR 1.11, 95% CI: 1.01–1.22, P=0.04) were factors independently associated with hospital mortality in acute respiratory failure patients without ARDS (Table 4), whereas baseline C_dyn (OR 1.01, 95% CI: 0.98–1.03, P=0.56) and actual respiratory rate (OR 0.98, 95% CI: 0.88–1.08, P=0.67) were not associated with mortality in the multivariable model.

Discussion

This study demonstrated high mortality (51%) among acute respiratory failure patients without ARDS, especially in patients who had severe organ dysfunction [median SOFA score was 10 (IQR, 7–13)]. The majority of patients in our cohort had mild hypoxemia on day 1 because most of them were intubated due to septic shock from extrapulmonary infections, and only 37% had pneumonia. There were differences in some parameters in baseline characteristics, treatment, respiratory variables, and complications between the surviving group and the deceased group. However, the multivariable analysis disclosed only higher PIP during the first three days of ICU admission, the presence of cerebrovascular disease, and the higher SOFA score at admission were associated with 28-day mortality.

The mortality of non-ARDS patients in our study was 51%, which was higher than that reported in the other non-ARDS cohorts (21–35%) (14-18). This could be explained by the finding that more organ dysfunction was found in our study. As noted in Table 1, the median SOFA score for overall patients was 10 (IQR, 7–13), 8 (IQR, 6–12) for alive, and 11 (IQR, 8–14) for dead patients, while median total SOFA score for patients in the prior non-ARDS cohorts was 6 (IQR, 4–9) (14,15). Moreover, multivariate analysis showed an association between higher admission SOFA scores and 28-day mortality. The predicted mortality rate by SOFA score was correlated with the mortality rate of non-ARDS patients in our population and the previous studies (14,15); the SOFA score of 10 was related to an approximate 50% mortality rate in critically ill patients, while the SOFA score of 6 predicted a 20–30% mortality rate (11); The other two non-ARDS cohorts did not report SOFA scores; one used acute physiology scores (17), and one used SAPS scores (18), which did not directly reflect the severity of organ dysfunction. Since most patients in this study had mild hypoxemia (Table 1), high SOFA scores should mainly result from non-respiratory organ dysfunctions. Achieving early definitive treatment for the primary disease and providing comprehensive organ support therapy is crucial, in addition to ventilator adjustment, for managing acute respiratory failure patients without ARDS.

Concerning MV strategy, we found that higher average PIP during the first 3 days of ICU admission was the only factor associated with increased mortality. Two prior cohort studies on non-ARDS patients reported similar results. In the PRoVENT trial, maximum airway pressure (P_max) was the only ventilatory variable independently associated with hospital mortality (14). Another study by Sahetya et al. demonstrated that plateau (P_plat) and driving pressure, the difference between P_plat and PEEP, were associated with higher mortality in these patients (15). On the contrary, some studies found no relationship between the plateau and driving pressure and mortality in non-ARDS patients, but in these studies, only a minority of patients (4–7%) were intubated due to pulmonary conditions (17,18). The discrepancy in patient characteristics may be attributable to different results, as 37% of patients in our cohort had pneumonia, and 28–50% of patients in the PRoVENT trial and Sahetya et al. were intubated for respiratory conditions (14,15).

The definition of PIP in our study differs from the P_max in the PRoVENT study and P_plat in Sahetya et al. The PIP in this study was the maximum airway pressure of both pressure-controlled and volume-controlled ventilation, not plateau pressure measured by end-inspiratory hold maneuver. Most patients (86%) in our cohort received pressure-controlled ventilation. The P_max in the PRoVENT study was defined as PIP in pressure-controlled ventilation and plateau pressure in volume-controlled ventilation (14). The definition of P_plat in the Sahetya et al. study was similar to that of P_max in the PRoVENT trial (14,15). We used the average values of respiratory variables over the first three days of ICU admission rather than a single data point to represent the actual ventilator parameters. The PIP in our study was attributed to the ΔP used to ventilate the patients because the PEEP levels were similar in both groups, so we omitted the ΔP in the multivariate analysis. We hypothesized that the baseline C_dyn might influence the PIP level; thus, we included the baseline C_dyn in the multivariable model, and the result confirmed that ventilating with higher PIP increased the risk of 28-day mortality independent of the baseline C_dyn value. Notably, deceased patients were more likely to experience worsening lung injury on day 7. This may be due to ventilator-induced lung injury caused by high PIP levels. Thus, carefully ventilating patients with the lowest possible PIP to achieve an acceptable gas exchange and patient-ventilator synchrony might be an appropriate MV strategy to improve outcomes of patients without ARDS who require MV.

We found an association between a prior history of cerebrovascular diseases and 28-day mortality in non-ARDS patients requiring MV (Table 4). We hypothesized that patients with a previous history of cerebrovascular diseases may have poor functional status, which leads to limited therapeutic effort; the findings in the PRoVENT trial support this hypothesis. The partially dependent functional status of the patients before admission was independently associated with in-hospital mortality in the multivariate analysis of the PRoVENT study (14).

This study had some limitations. First, our study collected data from a tertiary care and academic hospital in Bangkok, Thailand. Due to a shortage of ICU beds, only the most severely ill patients in our hospital were transferred to the ICU. As a result, most of the patients in our study had severe organ dysfunction. This might limit the generalizability of our study results to other hospitals with different circumstances. Second, we obtained hourly MV parameters from the nurse records. Variations in respiratory parameters within each hour were not taken into account. Third, we could not obtain plateau pressure and static driving pressure, the difference between plateau pressure and PEEP, since they are not regularly monitored for non-ARDS patients in our unit. Nonetheless, we believe that PIP is a more practical measure, as it is constantly displayed on the ventilator screen and does not require any special maneuvers. Fourth, despite most patients in our cohort having septic shock as a principal diagnosis, the data on vasoactive drug use was not available in our record. However, the use of vasoactive drugs is a part of the SOFA scores used to assess the disease severity, and we also found an association between higher SOFA scores and 28-day mortality in the multivariate analysis results. Finally, only 13% of patients (n=17) in this study had a history of strokes before admission. And we did not have data on their baseline functional status. Hence, further research is required to confirm an association between prior history of cerebrovascular diseases and mortality in non-ARDS patients.

Conclusions

Mortality was high in acute respiratory failure patients without ARDS. Factors associated with 28-day mortality included higher average PIP during the first three days of ICU admission, higher baseline SOFA scores, and underlying cerebrovascular diseases. Thus, carefully ventilating with the lowest possible PIP to achieve an acceptable gas exchange and patient-ventilator synchrony might be a suitable MV strategy to improve the outcomes of these patients. Further randomized controlled studies should be performed to confirm this hypothesis.

Acknowledgments

We thank Khemajira Karaketklang for helping with the statistical analysis.

Funding: None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-24-58/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-24-58/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-24-58/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-24-58/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This study was approved by the Human Research Protection Unit of the Faculty of Medicine Siriraj Hospital, Mahidol University (certificate of approval No. Si 546/2017). Written informed consent was obtained from all patients or their legal guardians if the participants could not provide the consent.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Mahmood SS, Pinsky MR. Heart-lung interactions during mechanical ventilation: the basics. Ann Transl Med 2018;6:349. [Crossref] [PubMed]
2. Grasselli G, Calfee CS, Camporota L, et al. ESICM guidelines on acute respiratory distress syndrome: definition, phenotyping and respiratory support strategies. Intensive Care Med 2023;49:727-59. [Crossref] [PubMed]
3. Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000;342:1301-8. [Crossref] [PubMed]
4. Combes A, Hajage D, Capellier G, et al. Extracorporeal Membrane Oxygenation for Severe Acute Respiratory Distress Syndrome. N Engl J Med 2018;378:1965-75. [Crossref] [PubMed]
5. Writing Group for the Alveolar Recruitment for Acute Respiratory Distress Syndrome Trial (ART) Investigators. Effect of Lung Recruitment and Titrated Positive End-Expiratory Pressure (PEEP) vs Low PEEP on Mortality in Patients With Acute Respiratory Distress Syndrome: A Randomized Clinical Trial. JAMA 2017;318:1335-45. [Crossref] [PubMed]
6. Amato MB, Meade MO, Slutsky AS, et al. Driving pressure and survival in the acute respiratory distress syndrome. N Engl J Med 2015;372:747-55. [Crossref] [PubMed]
7. Costa ELV, Slutsky AS, Brochard LJ, et al. Ventilatory Variables and Mechanical Power in Patients with Acute Respiratory Distress Syndrome. Am J Respir Crit Care Med 2021;204:303-11. [Crossref] [PubMed]
8. Serpa Neto A, Deliberato RO, Johnson AEW, et al. Mechanical power of ventilation is associated with mortality in critically ill patients: an analysis of patients in two observational cohorts. Intensive Care Med 2018;44:1914-22. [Crossref] [PubMed]
9. Neto AS, Barbas CSV, Simonis FD, et al. Epidemiological characteristics, practice of ventilation, and clinical outcome in patients at risk of acute respiratory distress syndrome in intensive care units from 16 countries (PRoVENT): an international, multicentre, prospective study. Lancet Respir Med 2016;4:882-93. [Crossref] [PubMed]
10. ARDS Definition Task Force. Acute respiratory distress syndrome: the Berlin Definition. JAMA 2012;307:2526-33. [PubMed]
11. Ferreira FL, Bota DP, Bross A, et al. Serial evaluation of the SOFA score to predict outcome in critically ill patients. JAMA 2001;286:1754-8. [Crossref] [PubMed]
12. Kangelaris KN, Calfee CS, May AK, et al. Is there still a role for the lung injury score in the era of the Berlin definition ARDS? Ann Intensive Care 2014;4:4. [Crossref] [PubMed]
13. PASW Statistics for Windows, Version 18.0. Chicago: SPSS Inc., 2009.
14. Simonis FD, Barbas CSV, Artigas-Raventós A, et al. Potentially modifiable respiratory variables contributing to outcome in ICU patients without ARDS: a secondary analysis of PRoVENT. Ann Intensive Care 2018;8:39. [Crossref] [PubMed]
15. Sahetya SK, Mallow C, Sevransky JE, et al. Association between hospital mortality and inspiratory airway pressures in mechanically ventilated patients without acute respiratory distress syndrome: a prospective cohort study. Crit Care 2019;23:367. [Crossref] [PubMed]
16. Pisani L, Algera AG, Serpa Neto A, et al. Epidemiological Characteristics, Ventilator Management, and Clinical Outcome in Patients Receiving Invasive Ventilation in Intensive Care Units from 10 Asian Middle-Income Countries (PRoVENT-iMiC): An International, Multicenter, Prospective Study. Am J Trop Med Hyg 2021;104:1022-33. [Crossref] [PubMed]
17. Lanspa MJ, Peltan ID, Jacobs JR, et al. Driving pressure is not associated with mortality in mechanically ventilated patients without ARDS. Crit Care 2019;23:424. [Crossref] [PubMed]
18. Schmidt MFS, Amaral ACKB, Fan E, et al. Driving Pressure and Hospital Mortality in Patients Without ARDS: A Cohort Study. Chest 2018;153:46-54. [Crossref] [PubMed]