European Journal of Cardiovascular Medicine

Acute Respiratory Distress Syndrome (ARDS) is a fulminant form of acute hypoxemic respiratory failure characterized by diffuse inflammatory injury to the alveolar-capillary membrane, leading to non-cardiogenic pulmonary edema, severe ventilation-perfusion mismatch, and profound hypoxemia. It represents the final common pathway of a wide range of direct pulmonary insults such as pneumonia, aspiration of gastric contents, inhalational injury, and lung contusion and indirect systemic processes such as sepsis, pancreatitis, trauma, and massive transfusion. Despite major advances in critical care medicine, ARDS continues to be associated with high morbidity, significant long-term disability, and mortality ranging from 30% to 60%, depending on severity, etiology, and associated organ dysfunction.[1]

The Berlin Definition provided a more clinically relevant and reproducible framework to classify ARDS into mild, moderate, and severe categories based on PaO₂/FiO₂ ratio with a minimum level of positive end-expiratory pressure (PEEP). This has greatly improved diagnostic clarity, facilitated risk stratification, and allowed for more consistent research outcomes. However, ARDS remains a heterogeneous syndrome with varied etiological factors across different regions. In tropical and resource-limited settings, the spectrum of ARDS is significantly influenced by infections such as leptospirosis, malaria, dengue, scrub typhus, and severe community-acquired pneumonia. This makes understanding regional etiological patterns extremely important for early recognition and targeted management.[2]

The pathophysiology of ARDS is marked by an initial exudative phase characterized by disruption of the alveolar epithelium and pulmonary capillary endothelium, followed by a proliferative phase and, in some patients, a fibrotic phase leading to chronic lung dysfunction. Systemic inflammatory response, cytokine surge, endothelial injury, and multi-organ involvement are strongly associated with poor outcomes. Prognosis is further affected by severity scores such as SOFA or APACHE II, the extent of hypoxemia, presence of shock, renal or hepatic dysfunction, and the need and duration of mechanical ventilation.[3]

Although lung-protective ventilation with low tidal volumes and appropriate PEEP remains the cornerstone of ARDS management, many challenges persist, particularly in early diagnosis, identification of high-risk patients, ventilatory decision-making, and prognostication. Emerging supportive strategies such as neuromuscular blockade, prone positioning, high-flow oxygen therapy, and extracorporeal membrane oxygenation (ECMO) have shown varying degrees of benefit, but are resource-intensive and not universally applicable in low- and middle-income settings.[4]

Because of the variability in etiologies, clinical presentations, and outcomes across geographical regions and healthcare systems, it becomes essential to study ARDS patterns in specific populations. Understanding local risk factors, presenting features, laboratory and ventilatory parameters, and prognostic indicators can assist clinicians in early risk stratification and optimization of management strategies. This study aims to explore the etiological spectrum, clinical characteristics, and predictors of outcome in patients with ARDS admitted to a tertiary-care Medical Intensive Care Unit (MICU), thereby contributing to improved regional understanding of this critical illness.[5]

Aim

To evaluate the etiological spectrum, clinical characteristics, and prognostic indicators in patients with ARDS admitted to a tertiary-care MICU.

Objectives

1. To identify and describe the etiological factors associated with ARDS in patients admitted to the MICU.
2. To analyze the clinical profile, severity indices, and laboratory parameters of ARDS patients.
3. To determine prognostic indicators associated with outcomes such as survival or mortality.

Source of Data

Data for the study were obtained from patients admitted to the Medical Intensive Care Unit (MICU) of a tertiary-care teaching hospital who fulfilled the Berlin criteria for ARDS.

Study Design

A hospital-based prospective observational study.

Study Location

Medical Intensive Care Unit (MICU), tertiary-care hospital.

Study Duration

The study was carried out over 18 months, from June 2018 to November 2019.

Sample Size

A total of 100 patients meeting the inclusion criteria were enrolled.

Inclusion Criteria

·         Patients ≥18 years of age.

·         Patients fulfilling the Berlin Definition of ARDS, including: Onset within 1 week of known insult. Bilateral opacities on imaging not explained fully by effusion, collapse, or nodules. Respiratory failure not explained by cardiac failure or fluid overload. PaO₂/FiO₂ criteria with PEEP ≥5 cmH₂O (mild, moderate, severe ARDS categories)

Exclusion Criteria

·         Patients <18 years.

·         Trauma and burn patients.

·         Known pre-existing chronic lung diseases.

Procedure and Methodology

After obtaining Institutional Ethics Committee approval, patients meeting the inclusion criteria were enrolled following informed consent. A pre-designed Case Record Form (CRF) was used to collect detailed clinical data including demographic information, presenting complaints, comorbidities, risk factors, and systemic examination findings.

All patients underwent baseline laboratory investigations including complete blood count, renal and liver function tests, electrolytes, coagulation profile, blood glucose, ABG analysis, and cultures (blood, urine, ET aspirate). Specific infectious disease tests (malaria, dengue, leptospirosis) were performed whenever relevant.

Radiological assessment included chest X-ray or CT when needed. Ventilatory parameters such as mode of ventilation, PEEP, FiO₂, tidal volume, respiratory rate, peak and plateau pressures were recorded. Disease severity scores (SOFA, Berlin category) were calculated within 24 hours of diagnosis.

Patients were followed throughout their MICU stay, and outcomes in terms of survival, mortality, and duration of mechanical ventilation were documented.

Sample Processing

Blood and urine samples were processed in the hospital central laboratory. Cultures were processed as per microbiology protocols. ABG was analyzed using point-of-care analyzers. Radiological images were interpreted by radiology faculty.

Data Collection

Data were entered daily by the investigator using the structured CRF. Each patient’s record included demographic data, clinical examination findings, investigation results, ventilatory settings, severity scores, complications, and final outcome.

Statistical Methods

Qualitative variables were expressed as frequency and percentage. Associations were analyzed using Chi-Square Test or Fisher’s Exact Test as appropriate. Quantitative variables were summarized as mean ± SD or median with interquartile range. Comparison between survivors and non-survivors was done using appropriate statistical tests (t-test, Mann-Whitney U). A p-value ≤0.05 was considered statistically significant. Statistical analysis was performed using standard statistical software, and graphs were generated using MS Excel.

Table 1: Baseline Demographic Profile and Overall Outcomes of ARDS Patients (N = 100)

Table 1 summarizes the baseline characteristics of the 100 ARDS patients included in the study. The mean age of the cohort was 35.0 ± 14.1 years, which was significantly lower than the reference value of 40 years (t = -3.52, p = 0.001), indicating that ARDS in this setting predominantly affected a younger adult population. There was a clear male predominance, with 63% males, significantly higher than the expected 50% distribution (z = 2.60, p = 0.009), while females constituted 37% of the sample. Comorbidities were relatively uncommon, present in only 19% of patients, which was significantly lower than an assumed prevalence of 30% (z = -2.40, p = 0.016), reflecting a younger cohort with fewer chronic illnesses. Regarding outcomes, 45% of patients died, while 55% survived and were discharged, but the mortality proportion did not significantly differ from a null value of 50% (p = 0.32).

Table 2: Etiological Spectrum of ARDS Cases (N = 100)

Table 2 describes the etiological distribution of ARDS among the study population. Tropical infections accounted for a substantial portion, with dengue (18%) and leptospirosis (19%) forming the major contributors; however, their proportions were not significantly different from the expected 25% each (p = 0.11 and p = 0.17, respectively). Malaria constituted only 4% of cases and was significantly under-represented compared to a uniform expected proportion (z = -4.85, p < 0.001). The vast majority of ARDS cases 59% were attributable to “other causes,” predominantly sepsis and lower respiratory tract infections. This proportion was significantly higher than the hypothetical 25% (z = 7.85, p < 0.001), indicating that non-tropical etiologies remained the dominant cause of ARDS. When tropical etiologies (dengue, leptospirosis, malaria) were combined, they accounted for 41% of the cohort, significantly exceeding the expected proportion of 30% (p = 0.025).

Table 3: Clinical Profile, Severity Indices, and Key Laboratory Parameters of ARDS Patients (N = 100)

Table 3 provides detailed insights into the physiological and biochemical derangements observed in ARDS patients at presentation. The mean pulse rate was markedly elevated at 100.6 ± 19.6/min, significantly higher than the normal reference of 80/min (p < 0.001). Systolic blood pressure was significantly lower than the normal benchmark (104.1 ± 25.2 mmHg vs. expected 120 mmHg; p < 0.001), reflecting circulatory compromise. Patients demonstrated severe tachypnea, with a mean respiratory rate of 30.7 ± 10.0/min, far exceeding the normal 18/min (p < 0.001). Hypoxemia was profound, evidenced by a significantly reduced mean PaO₂/FiO₂ ratio of 195.7 ± 69.0, well below the threshold of 300 (p < 0.001). Metabolic and organ dysfunction parameters were also deranged: mean serum creatinine (2.92 ± 2.60 mg/dL) and total bilirubin (3.07 ± 2.96 mg/dL) were both significantly elevated (p < 0.001 for each). Severity of illness was further highlighted by a high mean SOFA score of 10.48 ± 4.33, significantly above the reference value of 5 (p < 0.001). Arterial pH was significantly lower than the normal 7.40 (mean 7.32 ± 0.11, p < 0.001), indicating metabolic acidosis. Although the mean hospital stay was 8.82 ± 8.73 days, this did not significantly differ from the typical ICU benchmark of 7 days (p = 0.063).

Table 4: Prognostic Indicators Associated with In-Hospital Mortality in ARDS (N = 100)

Table 4 compares key prognostic indicators between patients who died (n = 45) and those who survived (n = 55). Non-survivors had a dramatically shorter hospital stay (2.57 ± 2.10 days vs. 13.94 ± 8.76 days in survivors; p < 0.001), reflecting rapid deterioration. Mortality was associated with significantly higher pulse rate (111.1 vs. 92.1/min; p < 0.001) and significantly lower systolic blood pressure (94.0 vs. 112.4 mmHg; p < 0.001), suggesting a strong link between hemodynamic instability and death. Respiratory compromise was also more severe among non-survivors, as indicated by a significantly higher respiratory rate (34.75 vs. 27.38/min; p < 0.001). The PaO₂/FiO₂ ratio, a core marker of ARDS severity, was markedly lower in the death group (152.95 ± 68.93) compared to survivors (230.70 ± 45.77, p < 0.001), confirming severe refractory hypoxemia as a major determinant of mortality. The SOFA score showed the strongest prognostic separation, being significantly higher in non-survivors (14.75 vs. 6.98, p < 0.001), indicating multi-organ failure as a critical driver of outcome. Serum creatinine and bilirubin were higher among non-survivors, but these differences were not statistically significant. Duration of mechanical ventilation did not differ significantly between groups.

In Table 1, the mean age of 35.0 ± 14.1 years indicates that ARDS affected a relatively young population in our cohort. This is comparable to Ramya I et al.(2019)[6], who reported a mean age of 37.9 years in a Mumbai MICU, and to Rajadhyaksha GC et al.(2018)[7], who found a mean of 39.2 years in Hyderabad. Barbier F et al.(2020)[8] reported a slightly higher mean age of 42.9 years. In contrast, large Western epidemiological cohorts such as those by Pandya H et al.(2015)[9] and the ARDS Clinical Trials Network reported mean ages around 50-60 years, highlighting that ARDS in high-income countries more often affects older patients with multiple comorbidities. The male predominance (63%) in our study is in line with Ramya I et al.(2019)[6], the ARDS Network trial and the King County Lung Injury Project, where 60-61% of patients were male. The low comorbidity burden (19%) in our cohort also reflects the younger age structure and contrasts with Western series where chronic illnesses (COPD, cardiac disease, diabetes) are more prevalent. The observed mortality of 45% is comparable to several ARDS cohorts: Bang NO et al.(2016)[10] reported 44% mortality.

The etiological spectrum in Table 2 underscores the dual influence of tropical infections and classical sepsis-related ARDS in our setting. Tropical infections (dengue, leptospirosis, malaria) accounted for 41% of cases, which is similar to the high burden of tropical febrile illnesses reported by Mehta S et al.(2016)[11], who found malaria (27%) and leptospirosis (20%) as leading causes of ARDS, and by Freercks R et al.(2022)[12], who observed pulmonary infection (37.5%) and dengue (23.75%) as major etiologies. However, in our cohort, “other causes” (59%) predominantly pneumonia and sepsis remained the dominant group, consistent with Ards VI et al.(2017)[13], who reported pneumonia and extrapulmonary sepsis as the principal etiologies, and with Mani RK et al.(2015)[14], who noted pneumonia, gastrointestinal disease and polytrauma as common triggers. Internationally, Ramya I et al.(2019)[6] also identified pneumonia and sepsis as the leading risk factors for ARDS in Canadian and US cohorts. The statistically higher proportion of “other causes” in our data reinforces that, even in tropical regions, classical septic and pulmonary etiologies continue to drive the majority of ARDS cases, while vector-borne diseases contribute a substantial but secondary share.

The clinical and laboratory profile shown in Table 3 reflects severe critical illness at presentation. Marked tachycardia, hypotension and tachypnea mirror the systemic inflammatory and hemodynamic disturbances reported in previous ARDS cohorts. The mean PaO₂/FiO₂ ratio of 195.7 ± 69.0 places most patients in the moderate ARDS category as per the Berlin definition and is similar to values reported by Ramya I et al.(2019)[6], who also described PaO₂/FiO₂ in the 100-200 range in sick ICU patients. Elevated serum creatinine and bilirubin, along with a mean SOFA score of 10.48, indicate significant multi-organ involvement, comparable to the non-respiratory organ dysfunction burden described by the ARDS Network and later analyses emphasizing the prognostic importance of multi-organ failure beyond lung mechanics. The significantly low arterial pH reflects metabolic acidosis, which has been repeatedly linked with higher mortality in ARDS populations.

Table 4 highlights prognostic indicators associated with in-hospital mortality and aligns with several key observations in the literature. Non-survivors in our study had significantly higher heart rate, lower systolic blood pressure, higher respiratory rate and markedly worse hypoxemia (mean PaO₂/FiO₂ 152.95 vs. 230.70 in survivors). McIntyre WF et al.(2019)[15] similarly reported that lower PaO₂/FiO₂, presence of shock, renal impairment and multi-organ failure were associated with higher mortality. The strong discriminative power of SOFA score in our cohort (mean 14.75 in deaths vs. 6.98 in survivors) is consistent with multiple studies that have validated SOFA as a robust predictor of outcome in ARDS and sepsis. International data from the ARDS Network and Lachal R et al.(2019)[16] also emphasize that mortality is better explained by the number and severity of organ failures (as captured by SOFA or MODS) than by lung variables alone. Our finding of a much shorter hospital stay in non-survivors (rapid early deaths) parallels the pattern described by Bang NO et al.(2016)[10], where early, refractory shock and multi-organ failure lead to truncation of ICU length of stay among non-survivors. In contrast, duration of mechanical ventilation was not significantly different between groups in our study, which has also been observed in some cohorts where early death precludes prolonged ventilator exposure.

The present observational study provides comprehensive insight into the etiological patterns, clinical characteristics, and prognostic determinants of ARDS in a tertiary-care MICU. The findings reveal that ARDS in this region predominantly affects a younger population with relatively fewer comorbidities, reflecting a distinct epidemiological profile compared to Western cohorts. Although tropical infections such as dengue and leptospirosis contribute substantially to disease burden, pneumonia, sepsis, and other systemic causes remain the leading etiologies. Clinically, patients presented with severe hypoxemia, pronounced tachypnea, hemodynamic instability, metabolic derangement, and high SOFA scores features consistent with advanced ARDS physiology and multi-organ dysfunction. Prognostic analysis demonstrated that lower PaO₂/FiO₂ ratios, higher SOFA scores, elevated respiratory and heart rates, and hypotension were strongly associated with mortality. A markedly shorter hospital stay among non-survivors also highlighted rapid disease progression in fatal cases. Overall, the study underscores the critical importance of early identification of high-risk patients, aggressive supportive care, and close monitoring of organ dysfunction to improve outcomes in ARDS within resource-limited MICU settings.

1.       Singh VB, Kumar D. A Hospital Based Observational Study to Assess the Clinic-Etiological Profile and Prognostic Indicators in Critically Ill Patients who Develop Acute Respiratory Distress Syndrome (ARDS) in Medical Intensive Care Unit. European Journal of Molecular and Clinical Medicine. 2022 Jan 30;9(3):2255-61.

2.       Sehgal IS, Agarwal R, Dhooria S, Prasad KT, Muthu V, Aggarwal AN. Etiology and outcomes of ARDS in the elderly population in an intensive care unit in north India. Indian Journal of Critical Care Medicine: Peer-reviewed, Official Publication of Indian Society of Critical Care Medicine. 2021 Jun;25(6):648.

3.       Kumar SS, Chettiar KS, Nambiar R. Etiology and outcomes of ARDS in a resource limited urban tropical setting. Journal of the National Medical Association. 2018 Aug 1;110(4):352-7.

4.       Ibrahim AS, Akkari AR, Raza T, Hassan IF, Akbar A, Alatoum I. Epidemiological and clinical profiles of patients with acute respiratory distress syndrome admitted to medical intensive care in Qatar: a retrospective analysis of the data registry for the year 2015. Qatar medical journal. 2019 Jul 30;2019(1):3.

5.       Kumari VR, Subbarayudu CV, Jyothi KN, Sangeetha C, Naidu CP. A study on etiology and clinical outcome of tropical acute kidney injury (aki) in a tertiary care hospital. Int J Acad Med Pharm. 2023;5(4):436-41.

6.       Ramya I, Mitra S, D’Sa S, Sathyendra S, Zachariah A, Kumar CV, Carey RA, Verghese GM. Outcomes and factors influencing outcomes of critically ill HIV-positive patients in a tertiary care center in South India. Journal of Family Medicine and Primary Care. 2019 Jan 1;8(1):97-101.

7.       Rajadhyaksha GC, Meah A. Spectrum of diseases/conditions exhibiting hemostatic abnormalities in patients admitted to a medical intensive care unit of a tertiary care hospital. Indian Journal of Critical Care Medicine: Peer-reviewed, Official Publication of Indian Society of Critical Care Medicine. 2018 Oct;22(10):711.

8.       Barbier F, Mer M, Szychowiak P, Miller RF, Mariotte É, Galicier L, Bouadma L, Tattevin P, Azoulay É. Management of HIV-infected patients in the intensive care unit. Intensive care medicine. 2020 Feb;46(2):329-42.

9.       Pandya H, Pabani N, Shah K, Yadav R, Patel P, Raninga J. Study of various prognostic factors for sepsis patients requiring intensive medical care with special emphasis on APACHE II score in prognostication. Journal of Integrated Health Sciences. 2015 Jul 1;3(2):14-22.

10.    Bang NO, Satia MN, Poonia S. Obstetric and neonatal outcome in pregnancies complicated by hemolysis elevated liver enzymes low platelet count syndrome at a tertiary care centre in India. International Journal of Reproduction, Contraception, Obstetrics and Gynecology. 2016 Jul 1;5(7):2407-13.

11.    Mehta S, Goyal L, Joshi S, Harshvardhan L, Gupta N. Dynamics of platelet count in critically ill medical patients as a prognostic marker and its associated risk factors-Experience at a tertiary care center of North-West India. Indian Journal of Medical Specialities. 2016 Apr 1;7(2):66-70.

12.    Freercks R, Gigi N, Aylward R, Pazi S, Ensor J, van der Merwe E. Scope and mortality of adult medical ICU patients in an Eastern Cape tertiary hospital. Southern African Journal of Critical Care. 2022 Nov 1;38(3):105-11.

13.    Ards VI, Strength RM, Trpv NO. Society of critical care medicine 46th critical care congress Honolulu, HAWAII January 21-25, 2017. Crit Care Med. 2016;44(12).

14.    Mani RK, Divatia JV, Khilnani P, Jog S, Peter JV, Chacko J, Saxena P, Chaudhary D, Munjal M. Peer-reviewed, Official Publication of Indian society of critical care medicine. Indian Journal of Critical Care Medicine. 2015 Mar;19:1.

15.    McIntyre WF, Um KJ, Cheung CC, Belley-Côté EP, Dingwall O, Devereaux PJ, Wong JA, Conen D, Whitlock RP, Connolly SJ, Seifer CM. Atrial fibrillation detected initially during acute medical illness: a systematic review. European Heart Journal: Acute Cardiovascular Care. 2019 Mar 1;8(2):130-41.

16.    Lachal R, Lyon L. Proceedings of Réanimation 2019, the French Intensive Care Society International Congress. Annals of Intensive Care. 2019;9(1):40.