Skip to main content
Download PDF

Impact of initial ventilation strategies on in-hospital mortality in sepsis patients: insights from the MIMIC-IV database

BMC Pulmonary Medicine volume 25, Article number: 147 (2025) Cite this article

Abstract

Background

This study evaluates the impact of different initial ventilation strategies on in-hospital mortality among sepsis patients.

Methods

We included hospitalized sepsis patients who underwent mechanical ventilation from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database and categorized them into groups based on their initial ventilation strategy: non-invasive ventilation (NIV) and invasive mechanical ventilation (IMV). The main endpoint analyzed was in-hospital mortality. A propensity score matching model was employed to address confounding factors, and Cox survival analysis was performed in the matched cohort. Subgroup analyses were conducted to evaluate population heterogeneity.

Results

Among 19,796 patients who received mechanical ventilation, 10,073 (50.8%) initially received NIV. The analysis included 2935 matched pairs. Patients initially receiving NIV exhibited a higher survival rate (P = 0.009) and a 24% lower risk of in-hospital mortality compared to those initially receiving IMV (P < 0.001). Subgroup analysis indicated significant survival benefits with initial NIV for patients without malignant tumor (MT), or lower Sequential Organ Failure Assessment (SOFA) scores and higher PO2/FiO2.

Conclusion

Among sepsis patients, initial NIV is linked to increased in-hospital survival rates and reduced mortality risk, particularly in patients without concurrent MT, lower SOFA scores, and higher PO2/FiO2.

Peer Review reports

Introduction

Sepsis is characterized by a dysregulated immune response by the host to infection, which can cause life-threatening organ dysfunction [1]. Sepsis also poses a major challenge within the intensive care unit (ICU), characterized by high mortality rates, significant economic burdens, and detrimental effects on many patients such as disability, functional impairment, and decreased quality of life [2,3,4]. Additionally, respiratory failure frequently complicates sepsis, which may exacerbate lung damage and cause respiratory dysfunction. The resulting respiratory failure manifests as lactic acidosis, hypoxemia, or hypercapnia, guiding the selection of initial ventilation mode [5].

Non-invasive ventilation (NIV) and invasive mechanical ventilation (IMV) serve as primary methods for respiratory management in ICUs [6]. Compared to IMV, NIV offers unique advantages, such as reducing the risks associated with intubation, and sedation. However, NIV also entails risks of discomfort related to the mask, patient-machine asynchrony, and gastric insufflation. The primary risks of NIV in acute respiratory failure (ARF) include delayed intubation and increased aspiration [7,8,9]. Studies suggest that unsuccessful NIV application is a significant predictor of mortality in sepsis patients, and initial ventilation strategy is linked to patient outcomes [10]. Due to the limited sample size in existing studies and the low-quality evidence, there are no clear standards for identifying sepsis patients who might benefit from NIV, and inappropriate use of NIV can be harmful. Currently, guidelines lack specific recommendations on how to choose ventilation strategies [11,12,13]. Nevertheless, the role of mechanical ventilation in clinical practice remains indispensable, and researching the relationship between initial ventilation strategies and sepsis outcomes is vital for improving clinical diagnosis and treatment as well as patient prognoses.

Moreover, studies suggest that NIV is only effective in patients with combined cardiopulmonary dysfunction [14]. However, opinions vary regarding its benefits for critically ill patients [15] and immunocompromised cancer patients [16,17,18] NIV can also potentially cause renal damage [19,20,21]. Exploring the association between initial ventilation strategies and outcomes in sepsis patients with varying characteristics and identifying clinical populations that may benefit from specific ventilation strategies are equally crucial.

This research, utilizing the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database, aims to explore how initial ventilation strategies impact in-hospital mortality rates in sepsis patients and reports on the outcomes among patients with different characteristics.

Materials and methods

Study design

In this retrospective cohort analysis, we drew data from the expansive MIMIC-IV database developed by the Massachusetts Institute of Technology (MIT). It includes patient care records from over 60,000 individuals admitted to the ICU at Beth Israel Deaconess Medical Center (BIDMC) between 2008 and 2019 [22]. The project received approval from the Institutional Review Boards at both BIDMC and MIT. Access to the database was obtained after the successful completion of the Collaborative Institutional Training Initiative program. As MIMIC-IV contained only anonymized information, patient consent was exempt.

Study population, exposure, and outcomes

We selected patients from the MIMIC-IV database diagnosed with sepsis and receiving mechanical ventilation based on the Sepsis-3.0 guidelines, which require a confirmed or suspected infection and a Sequential Organ Failure Assessment (SOFA) score of 2 or higher [1]. Individuals younger than 18 years were excluded, and only data at the initial ICU stay was considered for patients with multiple admissions. The main exposure was the strategy of initial ventilation (NIV or IMV), and in-hospital mortality was the primary endpoint.

Variable extraction

We gathered baseline data within the first 24 h of ICU admission using Structured Query Language. The extracted data included demographic and clinical details such as age, gender, weight, race, and initial assessment scores—SOFA score, Simplified Acute Physiology Score II (SAPS II), and Glasgow Coma Scale (GCS). Vital signs included mean arterial pressure (MAP), heart rate, body temperature (in Celsius), and respiratory rate. We also collected laboratory measurements like white blood cell (WBC) count, hemoglobin levels, platelet count, lactate levels, eGFR, pH, partial pressures of oxygen (pO2), and carbon dioxide (pCO2), PO2/FiO2 within the first 24 h. For repeated measures within this timeframe, the earliest recorded values were used. The eGFR was calculated using the MDRD (Modification of Diet in Renal Disease) equation: eGFR = 186×(Scr − 1.154)×(age − 0.203)×(0.742 if female), where Scr denotes serum creatinine in mg/dL and age in years. The MDRD equation is validated for renal function assessment in chronic kidney disease (CKD). Data on comorbidities present at admission, such as respiratory failure, hypoxemia, hypercapnia, lactic acidosis, acute kidney failure (AKF), pneumonia, heart failure (HF), chronic kidney disease, cirrhosis, chronic obstructive pulmonary disease (COPD), stroke, and malignant tumor (MT), were also collected based on ICD-9 codes from the MIMIC-IV database. Variables with missing data exceeding 30% were excluded from analysis.

Statistical analysis

Categorical variables were presented as frequency and percentage (%). Continuous variables were reported either as means with standard deviation (SD) or medians with interquartile ranges (IQR), based on their distribution profiles. Continuous variables conforming to normal distribution were analyzed using the t-test. In contrast, those that did not follow a normal distribution, along with ordinal variables, were assessed using the Wilcoxon rank-sum test. Analysis of categorical variables was conducted using the Chi-square test or Fisher’s exact test.

Propensity score matching (PSM) was used to adjust for and equilibrate potential confounders, namely variables that were either clinically important or differed between baseline groups (p < 0.05) [23]. This study employed a one-to-one nearest neighbor matching, with the caliper width at 0.05. The effectiveness of PSM was evaluated by calculating the standardized mean differences (SMDs). Despite adjusting for measured confounders, PSM cannot account for unmeasured confounding factors. For the matched cohort, Kaplan-Meier (KM) survival curves and multivariable Cox proportional hazards models were utilized to report hazard ratios (HRs) and 95% confidence intervals (CIs) concerning the initial ventilation strategies and their outcomes. Confounders that remained significant post-matching were included in the multivariable Cox analysis. Furthermore, subgroup analyses were performed by splitting the matched population into different subgroups based on the presence of age, gender, HF, COPD, MT, AKF, CKD, cause of respiratory failure, and median of the PaO2/FiO2, quartiles of the SOFA score. Forest plots were used to report differences in the impact of NIV on mortality risk across different populations.

Data management and statistical analysis were conducted using R (v 4.4.0), with significance set at p < 0.05.

Results

Patient characteristics and outcomes before matching

An analysis of 23,316 sepsis patients revealed that 19,796 of them received mechanical ventilation. Of these, 10,073 (50.9%) initially received NIV, and 9,723 (49.1%) received IMV. On average, the included population was 67 years old (± 16.05), with a higher proportion of males (58.6%), and most being white(77.2%). The overall SOFA score was 5 (3,7); the SAPS II score was 38 (31,48). Compared to patients initially receiving IMV, those treated with NIV were older (68.54 ± 16.20), lighter in weight (81.82 ± 24.03), and had less severe illness as indicated by lower SOFA scores (5(3,7)) and SAPS II scores (37(29,46)) and GCS scores (14(13,15)), higher pH levels (7.37 ± 0.07), and lower serum lactate levels (2.14 ± 1.44). The IMV group demonstrated higher eGFR (79.33 ± 47.10) and PIO2/FIO2 (411.8 ± 259.3). Regarding comorbidities, compared to the IMV group, NIV patients showed an increased prevalence of HF (32.66% vs. 24.47%), COPD (8.84% vs. 6.35%), CKD(20.50% vs. 14.95%), AKF(42.85% vs. 32.39%), MT(18.27% vs. 13.94%), and hypoxemia (12.78% vs. 10.81%), while stroke(8.74% vs. 9.48%), hypercapnia (2.71% vs. 2.5%), and lactic acidosis (19.50% vs. 18.99%) rates remained similar between groups. There were no statistically significant differences in vital signs within 24 h of admission between the groups (Table 1).

Table 1 Baseline characteristics in the original cohort
Full size table

In the unmatched cohort, the overall in-hospital mortality rate stood at 14.6% (2884/19796). 12.7% (1276/10073) of patients treated with NIV and 16.5% (1608/9723) of patients treated with IMV died in hospital. The incidence of switching from NIV to IMV was 23.1% (2328/10073), and among these patients, the in-hospital mortality was recorded at 20.1% (469/2328).

Characteristics and outcomes of patients after matching

After matching, 2935 pairs of patients initially treated with NIV (% of the initial NIV cohort) and those treated with IMV were included. Post-matching, the imbalance in covariates as measured by the SMD significantly decreased, indicating effective matching. No difference in respiratory rate and SPO2 was observed between the groups before matching; after matching, the NIV group exhibited a lower respiratory rate (19.88 ± 4.20) with a significant difference (SMD: 0.192, P < 0.001) and a lower Spo2 (96.88 ± 2.44) with a significant difference (SMD:0.368, P < 0.001). After matching, the SAPS II score (42.10 ± 14.61; SMD: 0.054; P = 0.031) differed between groups (Table 2).

Table 2 Baseline characteristics in the matched cohort
Full size table

The overall mortality rate among the matched patients was 15.8% (928/5870), and the NIV group experienced a lower in-hospital mortality rate of 15.1% (444/2935) compared to 16.5% (484/2935) in the IMV group. The incidence of switching from NIV to IMV post-matching was 35% (1029/2935), with an in-hospital mortality rate of 21.1% (218/1029) observed in this group.

KM curves and survival analysis

In the analysis of matched samples, the KM survival estimates demonstrated a notable disparity in survival between patients initially treated with NIV and those with IMV. The survival rate was notably higher in the initial NIV group (P = 0.009) (Fig. 1). Further analysis using a multivariable Cox proportional hazards model confirmed that the initial ventilation strategy significantly impacted in-hospital mortality in sepsis patients. Specifically, those initially treated with NIV exhibited a 24% reduction in the risk of in-hospital death compared to those receiving IMV (HR: 0.76, 95% CI: 0.67–0.87, P < 0.001) (Table 3).

Fig. 1
figure 1

Kaplan-Meier survival analysis curve for the in-hospital mortality of matched cohort. HR: Hazard Ratio, CI: Confidence Interval

Full size image
Table 3 Multifactor Cox proportional hazards model in the matched cohort
Full size table

Subgroup analysis

Subgroup analyses stratified by gender, age, HF, COPD, malignancy, CKD, AKF, hypoxemia, hypercapnia, lactic acidosis, SOFA score, and oxygenation index evaluated the association between initial ventilation mode and hospital mortality. The hazard ratios (HR) for in-hospital mortality were significantly different in subgroups younger than 67 years and subgroups of men (P < 0.05). However, there were no statistical differences in subgroups older than 67 years and women (P > 0.05). The association between initial ventilation pattern and in-hospital mortality was statistically significant regardless of HF (P < 0.05). In contrast, HR values for in-hospital mortality were statistically significant only in patients without MT, CKD, AKF, or COPD (P < 0.05). In addition, in the subgroup grouped by respiratory failure etiology, HR for in-hospital mortality was significant only in the subgroup without hypoxemia, hypercapnia, and lactic acidosis(P < 0.05). Finally, for patients with SOFA scores less than 4 and oxygenation index ≥ 241, there was a statistically significant association between initial ventilation mode and in-hospital mortality. (P < 0.05)

The interaction analysis showed that in subgroups based on gender, HF, COPD, CKD, AKF, hypoxemia, hypercapnia, or lactic acidosis, the initial ventilation mode had no significant impact on mortality risk (P > 0.05). However, in subgroups based on age, MT, SOFA score, and PO2/FiO2, NIV was significantly associated with a reduced mortality risk compared to IMV (P < 0.05). Particularly, NIV was closely associated with a lower mortality risk in patients with no MT 、lower SOFA scores, and higher PO2/FiO2. In contrast, in patients with MT、 higher SOFA scores, and lower PO2/FiO2, the difference in mortality risk between NIV and IMV disappeared, suggesting that other factors, aside from the initial ventilation mode, may influence the outcomes in these subgroups (Tables 4 and 5; Figs. 2 and 3).

Table 4 Subgroup analysis of initial ventilation mode based on demographics and comorbidities in sepsis patients
Full size table
Table 5 Subgroup analysis of initial ventilation mode based on oxygenation and disease severity in sepsis patients
Full size table
Fig. 2
figure 2

Forest plot of HRs for the primary endpoint in different subgroups of the matched cohort. HR: Hazard Ratio, CI: Confidence Interval, HF heart failure, MT Malignant tumor, COPD chronic obstructive pulmonary disease, CKD chronic kidney disease, AKF acute kidney failure, NIV non-invasive ventilation, invasive mechanical ventilation IMV

Full size image
Fig. 3
figure 3

Forest plot of HRs for the primary endpoint in different subgroups of the matched cohort. HR: Hazard Ratio, CI: Confidence Interval, SOFA Sequential Organ Failure Assessment, PO2 partial pressure of oxygen, FiO2 Fraction of Inspired Oxygen, SOFA quartile Q1:<4; Q2:4–5; Q3:6–8; Q4:≥9, NIV non-invasive ventilation, invasive mechanical ventilation IMV

Full size image

Discussion

Among approximately 20,000 sepsis patients who received mechanical ventilation included in this retrospective analysis, nearly half were initially treated with NIV. These patients had milder conditions upon admission, lower SOFA scores, and a higher prevalence of COPD and HF. Following PSM, the survival analysis highlighted a decreased mortality risk in patients who started with NIV rather than IMV, especially among patients without MT, or lower SOFA scores, and higher PO2/FiO2.

The benefit of using NIV to treat sepsis patients remains controversial, and most mechanical ventilation recommendations for sepsis patients are derived from Acute Respiratory Distress Syndrome (ARDS) trials. The LUNG SAFE study, which included 2,813 patients with ARDS, indicated that 15% of ARDS patients used NIV, with varying rates of failure and mortality based on ARDS severity [15]. Ferrer et al. found that NIV not only decreased the need for intubation but also reduced occurrences of septic shock and ICU mortality when compared to high-concentration oxygen therapy in cases of severe acute hypoxemic respiratory failure (AHRF) [24]. Similarly, a multicenter randomized controlled trial (RCT) involving 40 patients with acute lung injury demonstrated lower intubation rates and mortality for those treated with NIV versus those receiving oxygen therapy alone [25]. Honrubia et al. observed that while NIV did not significantly reduce mortality rates compared to IMV in patients with ARF, it did lead to fewer intubations and complications [26]. Antonelli et al. demonstrated that NIV was comparable to invasive methods in enhancing gas exchange for patients with AHRF and could avoid endotracheal intubation and reduce the occurrence of ventilator-associated pneumonia [27]. A retrospective examination of pneumonia patients revealed that initial use of NIV was significantly associated with higher hospitalization survival in comparison to IMV but was only applicable to patients with cardiopulmonary diseases, and patients for whom NIV failed had a high in-hospital mortality rate [14]. The reasons behind this association may be that NIV is typically used at an earlier stage of disease intervention and control, helping to reduce complications, promote patient recovery, and facilitate early activity. These constitute the advantages of NIV in the management of sepsis [28]. The use of NIV may imply an earlier intervention in respiratory support, especially before patients reach the severe respiratory failure stage requiring IMV. This early intervention could help control the condition more quickly, prevent further deterioration, and thus reduce the mortality risk [29, 30]. Mechanical ventilation supports breathing and gas exchange, either partially or fully, with the primary aim of maintaining adequate alveolar ventilation and arterial oxygen levels, thus preventing respiratory acidosis and hypoxemia [9]. Compared to IMV, NIV appears to cause less interference with airway protection and defense [31]. Patients on IMV often require sedatives and muscle relaxants to aid ventilation, which can suppress their ability to breathe autonomously and cough, thereby increasing the risk of pneumonia and other complications [32, 33]. In contrast, patients on NIV retain some ability to breathe autonomously and expectorate, reducing the occurrence of complications. Patients using NIV, by avoiding endotracheal intubation and the discomfort and restrictions associated with mechanical ventilation, may engage more readily in early activity and rehabilitation exercises. These activities help improve the patient’s cardiopulmonary function, muscle strength, and overall health status, thereby accelerating the recovery process and decreasing mortality risk [34, 35]. Furthermore, by avoiding invasive procedures such as tracheal intubation, NIV lowers the risk of discomfort and delirium by maintaining patient alertness and interaction with their surroundings, which enhances treatment compliance and rehabilitation outcomes, thus lowering mortality [36].

Additionally, some studies report inconsistent conclusions. In research focused on initial ventilation strategy choices for older patients with pneumonia, compared to IMV, using NIV as the initial strategy did not result in lower in-hospital mortality rates; there was no clear harm signal from using NIV, but researchers suggest that NIV should be cautiously selected for such patients [37, 38]. Moreover, He et al. found that compared to standard oxygen therapy, while NIV treatment could improve the oxygenation index (PaO2/FIO2), it failed to decrease intubation rates in patients with early mild ARDS caused by pneumonia [39]. This inconsistency in studies may relate to the risk of lung injury exacerbated by NIV treatment in ARDS, where increased intrapulmonary pressure and promotion of inflammation are primary causes of damage [40]. Inappropriate use of NIV may adversely affect patient outcomes [41].

This study observed that sepsis patients initially treated with NIV exhibited higher in-hospital survival rates and lower mortality risk compared to those initially treated with IMV. This aligns with the experimental conclusions reported by Antonelli and Stefan et al. However, Valley et al. and Besen et al. did not observe such benefits, which might be attributed to several factors: First, our analysis included all ventilated sepsis patients rather than only older patients with pneumonia. Second, the experiments by Valley et al. excluded patients with COPD or HF, whereas our study did not exclude such patients, and the proportion of patients with these comorbidities was not high. Third, our findings focus on in-hospital mortality instead of 30-day mortality. Fourth, patients with severe conditions who were late in receiving restrictive intubation tended to be treated with NIV and died after discharge. Additionally, sepsis-induced respiratory failure manifests as hypoxemia, hypercapnia, or lactic acidosis, each requiring specific ventilation strategies. Prior studies lacked stratification by respiratory failure type, limiting ventilation mode selection guidance. Data extraction and matching in this study minimized the impact of this heterogeneity. It is noteworthy that while Stefan and Besen et al. believed the benefits of initial NIV might be due to comorbid cardiopulmonary diseases, our adjusted cohort subgroup analysis did not observe differences in survival benefits between subgroups with or without COPD or HF, possibly due to the inefficiency of subgroup analyses and the relatively few patients with HF or COPD in the matched cohort. Additionally, compared to IMV, initial NIV was linked to a decreased mortality risk only in patients without MT, possibly because patients with sepsis and concurrent MT have compromised immunity, a higher probability of NIV failure, and a poorer prognosis [42]. Differences in risk ratios were also observed in patients with different PO2/FiO2. The decrease in mortality risk with NIV in higher PO2/FiO2 patients was statistically significant [20]. In patients with lower oxygenation indices, disease severity, and subsequent NIV failure contributed to increased mortality risk. Furthermore, in different SOFA score subgroups, we observed that high SOFA scores were associated with nonsignificant survival benefits, and in patients with SOFA scores ≥ 4, no link was found between initial NIV and survival benefits, possibly due to increased rates of NIV failure, more aggressive and prolonged ventilator support, delayed intubation, and poorer prognosis in severely ill patients. This finding is consistent with previous studies [43, 44]. However, due to the inefficiency of subgroup analysis, further RCTs are needed to verify the relationship between initial ventilation strategy choices and outcomes in different sepsis populations.

Our study has several strengths. This study utilized a rich electronic medical record database with a sufficient sample size, enabling adjustments for disease severity at admission using validated indicators, including laboratory test results. Furthermore, PSM was utilized to adjust for multiple confounding factors. However, there are notable limitations. Firstly, it is a single-center retrospective study predominantly involving a Caucasian population, and its conclusions may not be applicable to other populations. Secondly, data collection through SQL queries limited data completeness and accuracy, introducing potential selection bias despite quality control measures. Thirdly, while PSM adjusted for multiple known confounders in this retrospective study, unmeasured factors remained potential confounders. Additionally, since this study is based on electronic medical records from routine clinical visits, there is an inherent risk of measurement errors in the indicators used. Fourthly, many clinical decisions, including patient symptoms and physician experience, are not recorded in electronic records, introducing potential bias - evidenced by the 35% NIV-to-IMV conversion rate. Lastly, our study only assessed the impact of the initial ventilation strategy on in-hospital mortality among sepsis patients; the broader effects of NIV on the prognosis of sepsis need to be validated by more comprehensive research. Future large-scale RCTs are still needed to further verify the relationship between NIV and sepsis.

Conclusion

In summary, in patients with sepsis, initial NIV compared to IMV is linked to elevated survival rates and a decreased risk of mortality, particularly in patients without MT, and those with higher PO2/FiO2 and a SOFA score of less than 4.

Data availability

The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

Abbreviations

MIMIC-IV:

Medical information mart for intensive care IV

NIV:

Non-invasive ventilation

IMV:

Invasive mechanical ventilation

MT:

Malignant tumor

SOFA:

Sequential organ failure assessment

ICU:

Intensive care unit

ARF:

Acute respiratory failure

BIDMC:

Beth israel deaconess medical center

SAPS II:

Simplified acute physiology score II

GCS:

Glasgow coma scale

MAP:

Vital signs included mean arterial pressure

in Celsius:

Heart rate; body temperature

AKF:

Admission, such as acute kidney failure

HF:

Heart failure

CKD:

Chronic kidney disease

COPD:

Cirrhosis, chronic obstructive pulmonary disease

FiO2 :

Fraction of inspired oxygen

eGFR:

Estimated glomerular filtration rate

References

  1. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA. 2016;315(8):801–10. https://doiorg.publicaciones.saludcastillayleon.es/10.1001/jama.2016.0287.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Guo DC, Zhuang HZ, Lin J, Zhi DY, Duan ML. Epidemiology of sepsis in Beijing from 2012 to 2018: analysis of hospital homepage databases derived from the Beijing public health system. BMC Public Health. 2022;22(1):2237. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12889-022-14725-1.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Machado FR, Cavalcanti AB, Bozza FA, Ferreira EM, Angotti Carrara FS, Sousa JL, et al. The epidemiology of sepsis in Brazilian intensive care units (the sepsis prevalence assessment database, SPREAD): an observational study. Lancet Infect Dis. 2017;17(11):1180–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/s1473-3099(17)30322-5.

    Article  PubMed  Google Scholar 

  4. Rudd KE, Johnson SC, Agesa KM, Shackelford KA, Tsoi D, Kievlan DR, et al. Global, regional, and National sepsis incidence and mortality, 1990–2017: analysis for the global burden of disease study. Lancet. 2020;395(10219):200–11. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/s0140-6736(19)32989-7.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Gotts JE, Matthay MA. Sepsis: pathophysiology and clinical management. BMJ. 2016;353:i1585. https://doiorg.publicaciones.saludcastillayleon.es/10.1136/bmj.i1585.

    Article  PubMed  Google Scholar 

  6. Jaber S, Bellani G, Blanch L, Demoule A, Esteban A, Gattinoni L, et al. The intensive care medicine research agenda for airways, invasive and noninvasive mechanical ventilation. Intensive Care Med. 2017;43(9):1352–65. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00134-017-4896-8.

    Article  PubMed  Google Scholar 

  7. Demoule A, Chevret S, Carlucci A, Kouatchet A, Jaber S, Meziani F, et al. Changing use of noninvasive ventilation in critically ill patients: trends over 15 years in Francophone countries. Intensive Care Med. 2016;42(1):82–92. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00134-015-4087-4.

    Article  PubMed  Google Scholar 

  8. Demoule A, Girou E, Richard JC, Taille S, Brochard L. Benefits and risks of success or failure of noninvasive ventilation. Intensive Care Med. 2006;32(11):1756–65. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00134-006-0324-1.

    Article  PubMed  Google Scholar 

  9. Pearson SD, Koyner JL, Patel BK. Management of respiratory failure: ventilator management 101 and noninvasive ventilation. Clin J Am Soc Nephrol. 2022;17(4):572–80. https://doiorg.publicaciones.saludcastillayleon.es/10.2215/cjn.13091021.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Mosier JM, Sakles JC, Whitmore SP, Hypes CD, Hallett DK, Hawbaker KE, et al. Failed noninvasive positive-pressure ventilation is associated with an increased risk of intubation-related complications. Ann Intensive Care. 2015;5:4. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13613-015-0044-1.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Evans L, Rhodes A, Alhazzani W, Antonelli M, Coopersmith CM, French C, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021. Crit Care Med. 2021;49(11):e1063–143. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/ccm.0000000000005337.

    Article  PubMed  Google Scholar 

  12. Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli M, Ferrer R, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock: 2016. Intensive Care Med. 2017;43(3):304–77. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00134-017-4683-6.

    Article  PubMed  Google Scholar 

  13. Rochwerg B, Brochard L, Elliott MW, Hess D, Hill NS, Nava S, et al. Official ERS/ATS clinical practice guidelines: noninvasive ventilation for acute respiratory failure. Eur Respir J. 2017;50(2). https://doiorg.publicaciones.saludcastillayleon.es/10.1183/13993003.02426-2016.

  14. Stefan MS, Priya A, Pekow PS, Lagu T, Steingrub JS, Hill NS, et al. The comparative effectiveness of noninvasive and invasive ventilation in patients with pneumonia. J Crit Care. 2018;43:190–6. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jcrc.2017.05.023.

    Article  PubMed  Google Scholar 

  15. Bellani G, Laffey JG, Pham T, Madotto F, Fan E, Brochard L, et al. Noninvasive ventilation of patients with acute respiratory distress syndrome. Insights from the LUNG SAFE study. Am J Respir Crit Care Med. 2017;195(1):67–77. https://doiorg.publicaciones.saludcastillayleon.es/10.1164/rccm.201606-1306OC.

    Article  PubMed  Google Scholar 

  16. Frat JP, Ragot S, Girault C, Perbet S, Prat G, Boulain T, et al. Effect of non-invasive oxygenation strategies in immunocompromised patients with severe acute respiratory failure: a post-hoc analysis of a randomised trial. Lancet Respir Med. 2016;4(8):646–52. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/s2213-2600(16)30093-5.

    Article  PubMed  Google Scholar 

  17. Gristina GR, Antonelli M, Conti G, Ciarlone A, Rogante S, Rossi C, et al. Noninvasive versus invasive ventilation for acute respiratory failure in patients with hematologic malignancies: a 5-year multicenter observational survey. Crit Care Med. 2011;39(10):2232–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/CCM.0b013e3182227a27.

    Article  PubMed  Google Scholar 

  18. Lemiale V, Mokart D, Resche-Rigon M, Pène F, Mayaux J, Faucher E, et al. Effect of noninvasive ventilation vs oxygen therapy on mortality among immunocompromised patients with acute respiratory failure: A randomized clinical trial. JAMA. 2015;314(16):1711–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1001/jama.2015.12402.

    Article  CAS  PubMed  Google Scholar 

  19. Darmon M, Clec’h C, Adrie C, Argaud L, Allaouchiche B, Azoulay E, et al. Acute respiratory distress syndrome and risk of AKI among critically ill patients. Clin J Am Soc Nephrol. 2014;9(8):1347–53. https://doiorg.publicaciones.saludcastillayleon.es/10.2215/cjn.08300813.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Drury DR, Henry JP, Goodman J, The effects of continuous, pressure breathing on kidney function. J Clin Invest. 1947;26(5):945–51. https://doiorg.publicaciones.saludcastillayleon.es/10.1172/jci101889.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Priebe HJ, Heimann JC, Hedley-Whyte J. Mechanisms of renal dysfunction during positive end-expiratory pressure ventilation. J Appl Physiol Respir Environ Exerc Physiol. 1981;50(3):643–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1152/jappl.1981.50.3.643.

    Article  CAS  PubMed  Google Scholar 

  22. Johnson AEW, Bulgarelli L, Shen L, Gayles A, Shammout A, Horng S, et al. MIMIC-IV, a freely accessible electronic health record dataset. Sci Data. 2023;10(1):1. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41597-022-01899-x.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Zhang Z. Propensity score method: a non-parametric technique to reduce model dependence. Ann Transl Med. 2017;5(1):7. https://doiorg.publicaciones.saludcastillayleon.es/10.21037/atm.2016.08.57.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Ferrer M, Esquinas A, Leon M, Gonzalez G, Alarcon A, Torres A. Noninvasive ventilation in severe hypoxemic respiratory failure: a randomized clinical trial. Am J Respir Crit Care Med. 2003;168(12):1438–44. https://doiorg.publicaciones.saludcastillayleon.es/10.1164/rccm.200301-072OC.

    Article  PubMed  Google Scholar 

  25. Zhan Q, Sun B, Liang L, Yan X, Zhang L, Yang J, et al. Early use of noninvasive positive pressure ventilation for acute lung injury: a multicenter randomized controlled trial. Crit Care Med. 2012;40(2):455–60. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/CCM.0b013e318232d75e.

    Article  PubMed  Google Scholar 

  26. Honrubia T, García López FJ, Franco N, Mas M, Guevara M, Daguerre M, et al. Noninvasive vs conventional mechanical ventilation in acute respiratory failure: a multicenter, randomized controlled trial. Chest. 2005;128(6):3916–24. https://doiorg.publicaciones.saludcastillayleon.es/10.1378/chest.128.6.3916.

    Article  PubMed  Google Scholar 

  27. Antonelli M, Conti G, Rocco M, Bufi M, De Blasi RA, Vivino G, et al. A comparison of noninvasive positive-pressure ventilation and conventional mechanical ventilation in patients with acute respiratory failure. N Engl J Med. 1998;339(7):429–35. https://doiorg.publicaciones.saludcastillayleon.es/10.1056/nejm199808133390703.

    Article  CAS  PubMed  Google Scholar 

  28. Ferreyro BL, Angriman F, Munshi L, Del Sorbo L, Ferguson ND, Rochwerg B, et al. Association of noninvasive oxygenation strategies with All-Cause mortality in adults with acute hypoxemic respiratory failure: A systematic review and Meta-analysis. JAMA. 2020;324(1):57–67. https://doiorg.publicaciones.saludcastillayleon.es/10.1001/jama.2020.9524.

    Article  CAS  PubMed  Google Scholar 

  29. Güldner A, Pelosi P, Gama de Abreu M. Spontaneous breathing in mild and moderate versus severe acute respiratory distress syndrome. Curr Opin Crit Care. 2014;20(1):69–76. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/mcc.0000000000000055.

    Article  PubMed  Google Scholar 

  30. Yoshida T, Uchiyama A, Matsuura N, Mashimo T, Fujino Y. The comparison of spontaneous breathing and muscle paralysis in two different severities of experimental lung injury. Crit Care Med. 2013;41(2):536–45. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/CCM.0b013e3182711972.

    Article  PubMed  Google Scholar 

  31. Girou E, Schortgen F, Delclaux C, Brun-Buisson C, Blot F, Lefort Y, et al. Association of noninvasive ventilation with nosocomial infections and survival in critically ill patients. JAMA. 2000;284(18):2361–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1001/jama.284.18.2361.

    Article  CAS  PubMed  Google Scholar 

  32. Alhazzani W, Belley-Cote E, Møller MH, Angus DC, Papazian L, Arabi YM, et al. Neuromuscular Blockade in patients with ARDS: a rapid practice guideline. Intensive Care Med. 2020;46(11):1977–86. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00134-020-06227-8.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Chanques G, Constantin JM, Devlin JW, Ely EW, Fraser GL, Gélinas C, et al. Analgesia and sedation in patients with ARDS. Intensive Care Med. 2020;46(12):2342–56. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00134-020-06307-9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Putensen C, Zech S, Wrigge H, Zinserling J, Stüber F, Von Spiegel T, et al. Long-term effects of spontaneous breathing during ventilatory support in patients with acute lung injury. Am J Respir Crit Care Med. 2001;164(1):43–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1164/ajrccm.164.1.2001078.

    Article  CAS  PubMed  Google Scholar 

  35. Sassoon CS, Zhu E, Caiozzo VJ. Assist-control mechanical ventilation attenuates ventilator-induced diaphragmatic dysfunction. Am J Respir Crit Care Med. 2004;170(6):626–32. https://doiorg.publicaciones.saludcastillayleon.es/10.1164/rccm.200401-042OC.

    Article  PubMed  Google Scholar 

  36. Grieco DL, Maggiore SM, Roca O, Spinelli E, Patel BK, Thille AW, et al. Non-invasive ventilatory support and high-flow nasal oxygen as first-line treatment of acute hypoxemic respiratory failure and ARDS. Intensive Care Med. 2021;47(8):851–66. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00134-021-06459-2.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Besen B, Park M, Ranzani OT. Noninvasive ventilation in critically ill very old patients with pneumonia: A multicenter retrospective cohort study. PLoS ONE. 2021;16(1):e0246072. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pone.0246072.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Valley TS, Walkey AJ, Lindenauer PK, Wiener RS, Cooke CR. Association between noninvasive ventilation and mortality among older patients with pneumonia. Crit Care Med. 2017;45(3):e246–54. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/ccm.0000000000002076.

    Article  PubMed  PubMed Central  Google Scholar 

  39. He H, Sun B, Liang L, Li Y, Wang H, Wei L, et al. A multicenter RCT of noninvasive ventilation in pneumonia-induced early mild acute respiratory distress syndrome. Crit Care. 2019;23(1):300. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13054-019-2575-6.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Hashemian SM, Mohajerani SA, Jamaati HR. Ventilator-induced lung injury. N Engl J Med. 2014;370(10):979–80. https://doiorg.publicaciones.saludcastillayleon.es/10.1056/NEJMc1400293.

    Article  CAS  PubMed  Google Scholar 

  41. Carteaux G, Millán-Guilarte T, De Prost N, Razazi K, Abid S, Thille AW, et al. Failure of noninvasive ventilation for de Novo acute hypoxemic respiratory failure: role of tidal volume. Crit Care Med. 2016;44(2):282–90. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/ccm.0000000000001379.

    Article  PubMed  Google Scholar 

  42. Adda M, Coquet I, Darmon M, Thiery G, Schlemmer B, Azoulay E. Predictors of noninvasive ventilation failure in patients with hematologic malignancy and acute respiratory failure. Crit Care Med. 2008;36(10):2766–72. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/CCM.0b013e31818699f6.

    Article  PubMed  Google Scholar 

  43. Antonelli M, Conti G, Moro ML, Esquinas A, Gonzalez-Diaz G, Confalonieri M, et al. Predictors of failure of noninvasive positive pressure ventilation in patients with acute hypoxemic respiratory failure: a multi-center study. Intensive Care Med. 2001;27(11):1718–28. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00134-001-1114-4.

    Article  CAS  PubMed  Google Scholar 

  44. Tonelli R, Fantini R, Tabbì L, Castaniere I, Pisani L, Pellegrino MR, et al. Early inspiratory effort assessment by esophageal manometry predicts noninvasive ventilation outcome in de Novo respiratory failure. A pilot study. Am J Respir Crit Care Med. 2020;202(4):558–67. https://doiorg.publicaciones.saludcastillayleon.es/10.1164/rccm.201912-2512OC.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

This work was supported by Elderly health research project of Jiangsu Province (LR2022003).

Author information

Authors and Affiliations

  1. Department of Intensive Care Medicine, The Second Affiliated Hospital of Nanjing Medical University, 121 Jiangjiayuan, Nanjing, Jiangsu, 210000, China

    Yuxin Lu, Jingtao Zhang, Wanglin Zhang, Kanlirong Wang, Ziang Li & Liqun Sun

  2. Department of Emergency Medicine, The Second Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu, 210000, China

    Hongwei Shi

Authors
  1. Yuxin Lu
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Jingtao Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Wanglin Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Hongwei Shi
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Kanlirong Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Ziang Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Liqun Sun
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

All authors contributed to the study conception and design. Writing - original draft preparation: [YL]; Writing - review and editing: [LS, JZ, WZ]; Conceptualization: [YL, LS, JZ, WZ]; Methodology: [YL, JZ, WZ, HS]; Formal analysis and investigation: [YL, KW, ZL, HS]; Funding acquisition: [LS]; Resources: [LS]; Supervision: [LS], and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Liqun Sun.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lu, Y., Zhang, J., Zhang, W. et al. Impact of initial ventilation strategies on in-hospital mortality in sepsis patients: insights from the MIMIC-IV database. BMC Pulm Med 25, 147 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12890-025-03610-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12890-025-03610-3

Keywords

BMC Pulmonary Medicine

ISSN: 1471-2466

Contact us