Sepsis is a prevalent condition in the emergency department (ED), with a global incidence of 12–15% and a case fatality rate of 16–20% (1–3). More than one-third of sepsis patients develop acute respiratory distress syndrome (ARDS), which is typically more severe than sepsis without ARDS, leading to poorer prognoses and higher mortality rates (2–5). Sepsis induced ARDS contributes significantly to critical care mortality worldwide, with rates exceeding 40% in resource-limited settings despite advances in management protocols.

ARDS is diagnosed based on the Berlin definition (6), which includes the onset of respiratory symptoms, chest imaging, and oxygenation levels. Chest radiographs have suboptimal interobserver reliability and are often under-recognized in clinical settings, while computed tomography (CT) scans are not widely available. Moreover, no biomarker for ARDS is currently endorsed in clinical practice (7). Recent advancements in ARDS diagnosis (7) have resulted in a new global definition that, for the first time, formally integrates lung ultrasound findings as diagnostic criteria, especially in resource-constrained environments where chest radiography or CT may be inaccessible or impractical. This global definition acknowledges particular ultrasound patterns such as bilateral B-line predominance, with or without consolidations and pleural abnormalities as acceptable imaging criteria for diagnosing ARDS when conventional radiographic imaging is unavailable. This development recognizes the increasing body of evidence supporting the accuracy of lung ultrasound and its specific applicability in low- and middle-income nations.

Point-of-care lung ultrasound (LUS) has developed into a swift, non-invasive, and portable bedside instrument for identifying pulmonary pathology. B-lines, which are vertical hyperechoic artifacts originating from the pleural line, signify elevated extravascular lung water and interstitial involvement. B-lines are a sensitive but nonspecific finding that could result from various etiologies, including community-acquired pneumonia, cardiogenic pulmonary edema, iatrogenic fluid overload due to aggressive resuscitation, chronic interstitial lung diseases, and acute respiratory distress syndrome (ARDS). In the emergency department, excessive fluid administration during initial resuscitation may frequently lead to the development of B-lines. Although there is no specificity for an individual diagnosis, the identification of B-lines in septic patients may yield significant prognostic implications. The presence of pulmonary pathology indicated by B-lines may identify patients necessitating closer monitoring, earlier respiratory support, or adjustments in fluid management strategies, independent of the underlying etiology. In resource-constrained environments lacking advanced imaging techniques and limited access to invasive monitoring, LUS serves as an effective alternative for bedside evaluation (8–10).

The prognostic value of B-lines identified through point-of-care lung ultrasound in septic patients arriving at the emergency department remains inadequately defined, especially in settings with limited resources. Determining the correlation between B-line presence and adverse outcomes may assist clinicians in risk stratification and inform management decisions in the absence of advanced diagnostic resources. This study aimed to assess the association between B-lines identified through point-of-care lung ultrasound and clinical outcomes in adult patients with sepsis presenting to the emergency department in a resource-constrained environment.

During the study period, 1,356 patients diagnosed with sepsis, as indicated by a qSOFA score of ≥2, were admitted to the emergency department. A total of 1,172 patients were excluded for the following reasons: 325 were transferred from other hospitals, 234 were transferred to another hospital, 131 had a history of heart failure, 152 had end-stage chronic kidney disease, 78 had active cancer, 87 had interstitial lung disease, 48 had acute myocardial infarction, 32 had cardiogenic shock, and 85 lacked ultrasound documentation.

Ultimately, 184 patients were enrolled in this study, with a mean age of 63.50 years, of whom 51.08% were men. Most patients were classified as being at triage level 2 (64.67%). The most common etiology of sepsis was pneumonia (84, 45.65%), followed by abdominal sepsis (52, 28.26%) and urosepsis (42, 22.82%). Patient characteristics are summarized in Table 1.

B-lines were detected in 30.43% of patients (n = 56), indicating the presence of pulmonary pathology with increased extravascular lung water (OR 4.2, 95% CI 1.1–15.8, p = 0.032).

In terms of clinical outcomes, in-hospital mortality was 16.07% (n = 9) in the B-lines group and 4.68% (n = 6) in the non-B-lines group (p < 0.001). Mechanical ventilation was required in 39.28% of patients in the B-lines group, compared to 7.03% in the non-B-lines group (p < 0.001). The median hospital length of stay was 11 days for patients with B-lines and 6 days for those without B-lines (p < 0.001). This was described in Table 2.

In the multivariable logistic regression analysis, the B-lines group exhibited a significantly higher hospital mortality rate (adjusted odds ratio, 4.6; 95% confidence interval, 2.5–5.5) compared to the non-B-lines group. The B-lines group exhibited a significantly higher rate of mechanical ventilation utilization compared to the non-B-lines group (adjusted odds ratio, 9.1; 95% confidence interval, 6.4–11.2) (Table 3).

B-lines are distinct ultrasound artifacts in lung ultrasound, originating at the pleural line, appearing as light ray-like projections that obscure background artifacts, and moving synchronously with lung sliding. The presence of B-lines indicates extravascular lung water, which is associated with various pathologies (17–24).

This study demonstrates that the presence of B-lines detected on LUS is strongly associated with adverse outcomes in adult septic patients presenting to the emergency department. Patients with B-lines had significantly higher rates of in-hospital mortality, mechanical ventilation requirement, ICU admission, and prolonged hospitalization compared to those without B-lines. These findings suggest that LUS can serve as a valuable bedside risk stratification tool in resource-limited emergency department settings, identifying septic patients who warrant closer monitoring and more aggressive supportive care.

The mortality rate in the B-lines group was 16.07%, compared to 4.68% in the non-B-lines group, suggesting that the presence of B-lines may be associated with an increased mortality risk in sepsis patients. Our findings are consistent with prior literature (25–27) demonstrating associations between lung ultrasound findings and adverse outcomes in critically ill patients. However, most previous studies have been conducted in well-resourced settings with access to comprehensive diagnostic modalities.

A significantly higher proportion of patients in the B-lines group required mechanical ventilation (39.28%) compared to the non-B-lines group (7.03%), indicating that B-lines are associated with greater respiratory impairment. Our study also found that the duration of mechanical ventilation was significantly longer in the B-lines group (median of 7 days) compared to the non-B-lines group (median of 4 days). The strong association between B-lines and mechanical ventilation requirement (OR 9.1, 95% CI 6.4–11.2) suggests that B-lines may identify patients with evolving respiratory failure who could benefit from earlier respiratory support interventions, though our study design cannot establish causation (28, 29).

Importantly, our study population had pneumonia as the most common source of sepsis (46.43% in B-lines group vs. 45.31% in non-B-lines group, p = 0.247). This underscores that B-lines are nonspecific findings that cannot distinguish between various causes of increased extravascular lung water. Nevertheless, our data demonstrate that B-line presence carries prognostic significance regardless of etiology.

The median hospital stay was significantly longer in the B-lines group (11 days) compared to the non-B-lines group (6 days), reinforcing the association between B-lines and increased illness severity, as well as prolonged hospitalization. This finding is consistent with previous studies indicating extended hospital stays in sepsis-related ARDS. Additionally, some studies have reported prolonged intensive care unit stays for these patients (29, 30).

The clinical importance of these findings pertains to the prognostic, rather than diagnostic, significance of B-lines in septic patients. B-lines are vertical hyperechoic artifacts that signify elevated extravascular lung water, arising from various causes such as pneumonia (the predominant cause of sepsis in our cohort at 46%), cardiogenic pulmonary edema, iatrogenic fluid overload due to resuscitation, chronic interstitial lung disease, or acute respiratory distress syndrome. Our study indicates that, independent of the underlying etiology, the presence of B-lines represents a high-risk cohort with significantly worse outcomes.

Our study confirms that lung ultrasound can serve as a valuable bedside tool for emergency physicians managing critically ill patients, particularly in resource-limited settings where advanced imaging modalities are unavailable. This aligns with studies on the utility of lung ultrasound in diagnosing and managing COVID-19, which suggested that lung POCUS could be a viable alternative to chest CT scans (31–33). Furthermore, our findings highlight the importance of identifying B-lines in the emergency department for patients with sepsis-related ARDS, as it influences clinical outcomes and is crucial for timely resuscitation and intensive care planning (34–36).

This study presents several significant strengths, including its implementation in a resource-limited emergency department setting where point-of-care ultrasound may exhibit optimal clinical utility and impact, the implementation of a standardized lung ultrasound protocol with clear, reproducible criteria for B-line positivity (≥3 B-lines in ≥2 bilateral lung zones), assessment of various clinically relevant results such as in-hospital mortality, necessity for mechanical ventilation, and ICU admission, as well as a sufficient sample size to identify significant associations with substantial effect sizes across multiple endpoints. However, several significant limitations deserve attention. Selection bias presents a significant issue since POCUS was conducted at the discretion of the treating physician instead of following a systematic protocol, potentially biasing our cohort toward more severely ill patients or those with heightened clinical suspicion for pulmonary pathology, thus restricting generalizability for a broader sepsis populations. The temporal ambiguity concerning the timing of ultrasound examinations in relation to fluid resuscitation constitutes a notable limitation; although POCUS was consistently conducted within 2 h of emergency department arrival, we were unable to systematically record the exact timing or volume of fluids administered prior, hindering our ability to differentiate B-lines indicative of pre-existing pulmonary pathology or early sepsis-induced lung injury from those caused by iatrogenic fluid overload. We are unable to define the specific cause of B-lines in individual patients unless they are due to pneumonia, sepsis-induced ARDS, cardiogenic pulmonary edema, or chronic interstitial illness. However, this nonspecificity does not inherently reduce prognostic value, as our findings indicate that B-line detection identifies high-risk patients irrespective of etiology, which may prove more immediately beneficial than an accurate etiological diagnosis in resource-constrained environments. Further methodological limitations encompass the single-center, retrospective design characterized by substantial exclusions, a modest sample size with a limited number of events, leading to low events-per-variable ratios beneath the standard threshold of 10, as well as numerous data deficiencies, including the absence of systematic comparisons with chest imaging, lack of microbiological data, undocumented ventilatory parameters, imprecise timing of interventions (diuretics, renal replacement therapy, vasopressors), and the absence of formal interobserver agreement assessments for B-line interpretation. These constraints highlight the necessity for validation via extensive, prospective, multicenter studies utilizing systematic ultrasound protocols with standardized timing in relation to resuscitation, thorough data collection, and blinded outcome evaluation to corroborate and expand our initial findings.

Our study indicates that B-lines identified on lung ultrasound are significantly correlated with adverse outcomes in adult septic patients arriving at the emergency department in a resource-constrained environment. B-lines are nonspecific indicators that may arise from various causes such as pneumonia, fluid overload, cardiogenic edema, or ARDS; their presence signifies high-risk patients who face markedly elevated rates of mortality, mechanical ventilation, ICU admission, and extended hospitalization. LUS serves as an efficient, swift, and non-invasive instrument for bedside risk assessment in the absence of advanced diagnostic resources. These findings argue for the incorporation of lung ultrasound in the preliminary evaluation of septic patients to inform clinical decision-making and resource distribution.