Sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection, a definition formalised in 2016 by the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3)¹. This definition replaced the earlier systemic inflammatory response syndrome (SIRS)-based criteria, recognising that the clinical syndrome of sepsis arises from a complex interplay of pathogen virulence factors and an exaggerated, poorly regulated host inflammatory and immune response that ultimately produces tissue hypoperfusion and multi-organ dysfunction^1,2. Septic shock, the most severe subset, is identified by the requirement for vasopressors to maintain a mean arterial pressure of 65 mmHg or greater together with a serum lactate level above 2 mmol/L despite adequate fluid resuscitation, and is associated with the highest mortality among ICU syndromes^1,3.

Globally, sepsis remains a major and largely under-recognised public health problem. The Global Burden of Disease Study estimated that in 2017 there were 48.9 million incident cases of sepsis worldwide, resulting in 11.0 million sepsis-related deaths, accounting for nearly one in five deaths globally⁴. Although age-standardised incidence and mortality have declined substantially since 1990, the burden remains disproportionately concentrated in low- and middle-income countries, particularly in sub-Saharan Africa and South Asia^4,5. This recognition prompted the World Health Assembly in 2017 to adopt a resolution calling on member states to improve the prevention, diagnosis, and clinical management of sepsis as a global health priority⁵.

Within the ICU specifically, sepsis and septic shock are among the most frequent reasons for admission and are associated with mortality rates that, despite decades of research, remain considerably higher than most other critical illnesses. Large multinational audits and national cohort studies have repeatedly demonstrated that a substantial proportion of all ICU admissions meet criteria for severe sepsis or septic shock within the first 24 hours of admission, with cardiovascular and respiratory dysfunction being the most commonly affected organ systems⁶. Single-centre studies from India and other resource-limited settings have similarly reported high in-ICU mortality, frequently exceeding 30% overall and approaching or exceeding 60% among patients who progress to septic shock, with gram-negative bacilli emerging as the predominant causative organisms⁷.

Several factors are believed to influence outcome in septic patients admitted to the ICU. These include the source and site of infection, the causative organism and its antimicrobial susceptibility, the time to initiation of appropriate antibiotics and source control, the severity of organ dysfunction at presentation as quantified by validated scoring systems such as the Acute Physiology and Chronic Health Evaluation (APACHE) II score and the Sequential Organ Failure Assessment (SOFA) score, and underlying patient comorbidities such as chronic kidney disease, diabetes mellitus, and cirrhosis^8,9. Notably, patients with underlying cirrhosis admitted with sepsis have been shown to have more than twice the odds of in-hospital death compared with non-cirrhotic septic patients, independent of organ dysfunction or haemodynamic status, underscoring the importance of host factors in determining prognosis⁹.

In response to the persistently high burden of sepsis-related mortality, the Surviving Sepsis Campaign has periodically updated international evidence-based guidelines for the early management of sepsis and septic shock, with the most recent 2021 iteration emphasising early fluid resuscitation, timely administration of broad-spectrum antimicrobials, judicious use of vasopressors, and structured post-ICU follow-up for survivors¹⁰. Despite these advances, considerable variability in clinical profile, microbial epidemiology, and outcome persists across regions and even across individual ICUs, reflecting differences in case-mix, resource availability, and adherence to bundled care^6,10.

Given this background, there remains a continued need for context-specific data describing the clinical characteristics, microbiological spectrum, and outcomes of patients with sepsis admitted to the ICU, particularly to identify locally relevant predictors of mortality that can inform triage, resource allocation, and quality-improvement initiatives. The present study was therefore undertaken to describe the clinical profile and outcomes of patients with sepsis admitted to a tertiary care intensive care unit, and to identify the factors independently associated with mortality in this population.

MATERIALS AND METHODS

Study design and setting

This was a prospective, observational, single-centre study conducted in the multidisciplinary medical-surgical intensive care unit of a tertiary care teaching hospital over a period of 12 months. The study was approved by the Institutional Ethics Committee, and written informed consent was obtained from each patient or their legally authorised representative prior to enrolment, in accordance with the Declaration of Helsinki.

Study population

All adult patients aged 18 years and above who were admitted to the ICU and who fulfilled the Sepsis-3 consensus criteria for sepsis — defined as a suspected or confirmed infection accompanied by an acute increase of 2 points or more in the Sequential Organ Failure Assessment (SOFA) score — within the first 24 hours of ICU admission were consecutively enrolled¹. Septic shock was defined according to the same consensus criteria as sepsis with persisting hypotension requiring vasopressors to maintain a mean arterial pressure of 65 mmHg or greater and a serum lactate level exceeding 2 mmol/L despite adequate fluid resuscitation¹.

Patients younger than 18 years, those with a documented do-not-resuscitate order at the time of admission, pregnant women, and patients with an ICU stay of less than 24 hours due to early transfer or discharge were excluded from the study. Patients in whom informed consent could not be obtained were also excluded.

Data collection

Data were collected using a structured proforma by trained investigators at the time of ICU admission and during the course of the ICU stay. Recorded variables included age, sex, presenting comorbidities (diabetes mellitus, hypertension, chronic kidney disease, chronic liver disease, chronic obstructive pulmonary disease, malignancy, and coronary artery disease), suspected and confirmed source of infection, and results of microbiological cultures obtained from blood, urine, respiratory secretions, and other relevant sites prior to initiation of antimicrobial therapy.

Severity of illness was quantified using the Acute Physiology and Chronic Health Evaluation (APACHE) II score and the Sequential Organ Failure Assessment (SOFA) score, both calculated using the worst physiological and laboratory parameters recorded within the first 24 hours of ICU admission. Organ support requirements, including invasive mechanical ventilation, vasopressor or inotropic support, and renal replacement therapy, were recorded prospectively throughout the ICU stay. The primary outcome was ICU mortality, defined as death occurring during the ICU stay; secondary outcomes included hospital mortality, ICU length of stay, hospital length of stay, and duration of mechanical ventilation.

Statistical analysis

Data were entered into a structured spreadsheet and analysed using standard statistical software. Continuous variables were expressed as mean ± standard deviation or median with interquartile range, as appropriate to their distribution, and were compared between survivors and non-survivors using the independent samples t-test or Mann-Whitney U test. Categorical variables were expressed as frequencies and percentages and compared using the chi-square test or Fisher’s exact test, as applicable. Variables found to be significant on univariate analysis (p<0.05) were entered into a multivariate binary logistic regression model to identify independent predictors of ICU mortality, with results expressed as adjusted odds ratios (AOR) with 95% confidence intervals (CI). A p-value of less than 0.05 was considered statistically significant throughout.

RESULTS

A total of 240 patients meeting Sepsis-3 criteria were enrolled over the study period. The baseline demographic and clinical characteristics of the study population are summarised in Table 1.

Table 1. Baseline demographic and clinical characteristics of the study population (n = 240)

The mean age of the study population was 58.4 ± 14.6 years, with a male predominance of 61.7%. Diabetes mellitus (38.3%) and hypertension (35.0%) were the most common pre-existing comorbidities, followed by chronic kidney disease (16.7%) and chronic liver disease (9.2%). The respiratory tract was the most frequent primary source of infection (38.3%), followed by the urinary tract (21.3%), intra-abdominal sources (17.1%), skin and soft tissue infections (10.0%), and bloodstream infections of unidentified primary focus (8.3%). Septic shock at the time of ICU admission or during the ICU stay was documented in 35.4% of the cohort.

Table 2. Microbiological profile of culture-positive patients (n = 158)

Microbiological cultures were positive in 158 of 240 patients (65.8%). Among culture-positive cases, gram-negative organisms predominated, accounting for 61.4% of isolates, while gram-positive organisms accounted for 26.6% and fungal isolates for 12.0%. Klebsiella pneumoniae (22.8%) and Escherichia coli (17.7%) were the most frequently isolated gram-negative organisms, followed by Pseudomonas aeruginosa (12.0%) and Acinetobacter baumannii (8.9%). Among gram-positive isolates, Staphylococcus aureus (13.3%, of which 60% were methicillin-resistant) and Enterococcus species (8.2%) were most common. Candida species, predominantly Candida tropicalis and Candida albicans, accounted for the fungal isolates, identified largely in patients with prolonged ICU stay and prior broad-spectrum antibiotic exposure.

Table 3. Severity scores, organ support, and ICU course in the study population

The mean APACHE II score at admission was 21.6 ± 7.8 and the mean SOFA score was 8.2 ± 3.6, reflecting a moderately to severely ill cohort. Mechanical ventilation was required in 52.5% of patients, vasopressor or inotropic support in 41.7%, and renal replacement therapy in 18.3%. The median ICU length of stay was 7 days (interquartile range 4–12 days) and the median hospital length of stay was 14 days (interquartile range 9–21 days). Among patients requiring mechanical ventilation, the median duration of ventilation was 6 days (interquartile range 3–10 days).

Table 4. Comparison of baseline and clinical variables between survivors and non-survivors

*Mortality-by-source figures represent in-group death proportions, not column percentages of the survivor/non-survivor totals; SD: standard deviation; IQR: interquartile range; OR: odds ratio; CI: confidence interval.

Overall ICU mortality in the study cohort was 34.2% (82 of 240 patients), and hospital mortality was 37.5% (90 of 240 patients). On univariate comparison, non-survivors were significantly older (63.7 ± 13.1 years vs. 55.8 ± 14.9 years, p<0.001), had higher APACHE II scores (26.4 ± 6.9 vs. 19.1 ± 7.2, p<0.001) and higher SOFA scores (10.6 ± 3.1 vs. 6.9 ± 3.0, p<0.001) at admission, and were more likely to have septic shock (58.8% vs. 21.2%, p<0.001), to require mechanical ventilation (76.8% vs. 40.5%, p<0.001), and to require renal replacement therapy (32.9% vs. 11.4%, p<0.001) compared with survivors. Mortality also varied significantly with the source of infection, being highest among patients with intra-abdominal sepsis (43.9%) and lowest among those with urinary tract infection (21.6%).

Table 5. Independent predictors of ICU mortality on multivariate logistic regression analysis

On multivariate logistic regression analysis adjusting for age, comorbidity burden, and source of infection, four variables emerged as independent predictors of ICU mortality: presence of septic shock (adjusted odds ratio [AOR] 4.82, 95% CI 2.41–9.64, p<0.001), SOFA score ≥ 8 at admission (AOR 3.65, 95% CI 1.87–7.13, p<0.001), requirement for mechanical ventilation (AOR 2.94, 95% CI 1.42–6.08, p=0.004), and age greater than 65 years (AOR 2.21, 95% CI 1.12–4.36, p=0.022). Chronic kidney disease showed a trend toward increased mortality risk but did not reach statistical significance after adjustment (AOR 1.78, 95% CI 0.89–3.56, p=0.103).

DISCUSSION

This prospective observational study describes the clinical profile and outcomes of 240 patients with sepsis admitted to a tertiary care ICU, and demonstrates an overall ICU mortality of 34.2%, rising to 58.8% among patients who developed septic shock. These figures are broadly consistent with previously reported mortality rates from comparable ICU cohorts. A large Brazilian single-centre study of 971 ICU admissions reported an overall mortality of 31.1%, with mortality of 22.6% among patients with severe sepsis and 64.8% among those with septic shock⁷, figures that closely parallel the shock-associated mortality observed in the present cohort. Similarly, the landmark Intensive Care National Audit and Research Centre case-mix programme database analysis of over 56,000 adult ICU admissions in England, Wales, and Northern Ireland found that more than one-quarter of all ICU admissions met severe sepsis criteria within the first 24 hours, with cardiovascular and respiratory dysfunction representing the most frequently affected organ systems⁶, a pattern mirrored in our cohort where the respiratory tract was the leading source of infection and respiratory and cardiovascular failure were the dominant modes of organ dysfunction necessitating mechanical ventilation and vasopressor support.

The microbiological pattern observed in this study, with a clear predominance of gram-negative organisms — particularly Klebsiella pneumoniae and Escherichia coli — accounting for nearly two-thirds of culture-positive isolates, is consistent with patterns reported from other ICU cohorts in similar healthcare settings, where gram-negative bacilli have been repeatedly identified as the most prevalent pathogens in sepsis^7,8. This finding has direct implications for empirical antimicrobial selection in resource-limited ICU settings, where gram-negative coverage with attention to local antibiograms should remain a cornerstone of initial therapy pending culture results, in line with recommendations of the Surviving Sepsis Campaign for early, broad-spectrum, source-appropriate antimicrobial therapy¹⁰. The substantial proportion of methicillin-resistant Staphylococcus aureus among gram-positive isolates and the emergence of Candida species in a subset of patients with prolonged ICU stay further underscore the importance of antimicrobial stewardship and judicious de-escalation once culture sensitivities become available.

The independent predictors of mortality identified in this study — septic shock, higher SOFA score, need for mechanical ventilation, and advanced age — are well recognised determinants of outcome in the sepsis literature and reinforce the discriminatory value of validated severity scores at the time of ICU admission^1,8. The SOFA score, which forms the basis of the Sepsis-3 definition itself, has consistently demonstrated superior predictive validity for in-hospital mortality compared with the older SIRS-based criteria across multiple large cohorts¹, a finding corroborated in the present study by the markedly higher mean SOFA score among non-survivors compared with survivors. Similarly, advancing age has been repeatedly associated with reduced physiological reserve and higher case fatality in sepsis, likely reflecting both diminished organ reserve and a higher prevalence of comorbid disease in older patients⁸.

It is notable that although chronic kidney disease showed a trend toward higher mortality in this cohort, it did not retain independent significance after adjustment for shock status and organ dysfunction severity, a pattern that diverges somewhat from data in other high-risk comorbid populations. For example, a large retrospective cohort study of nearly 8,000 ICU patients with sepsis found that the presence of underlying cirrhosis was independently associated with more than double the odds of in-hospital mortality, even after adjustment for comorbidities, organ dysfunction, and haemodynamic status⁹, suggesting that certain chronic organ-specific comorbidities may carry prognostic weight that is not fully captured by generic severity scores alone, and that comorbidity-specific risk stratification may have a complementary role alongside SOFA and APACHE II scoring.

From a global health perspective, the mortality observed in this study, while substantial, must be interpreted against the backdrop of the considerable worldwide burden of sepsis. The Global Burden of Disease Study estimated that sepsis was implicated in nearly one-fifth of all deaths worldwide in 2017, with the highest burden concentrated in low- and middle-income regions⁴, precisely the settings in which delayed presentation, limited microbiological diagnostic capacity, and constrained critical care resources may compound the inherent lethality of the syndrome. This underscores the continued relevance of the Surviving Sepsis Campaign’s emphasis on early bundled care — timely fluid resuscitation, prompt antimicrobial therapy, and judicious haemodynamic support — as a means of mitigating preventable mortality, particularly in resource-constrained ICU environments¹⁰.

This study has certain limitations that merit consideration. As a single-centre study, the findings may not be fully generalisable to ICUs with differing case-mix, resource availability, or microbiological epidemiology. The relatively modest sample size limited the power to detect independent associations for less common comorbidities and infection sources. Additionally, long-term outcomes beyond hospital discharge, including post-ICU functional status and 90-day mortality, were not assessed and represent an important area for future research.

CONCLUSION

Sepsis continues to be associated with substantial mortality among patients admitted to the intensive care unit, with outcomes strongly influenced by the presence of septic shock, severity of organ dysfunction at presentation, requirement for mechanical ventilation, and advanced age. Gram-negative organisms remain the predominant causative pathogens in this setting, reinforcing the need for locally informed empirical antimicrobial protocols. Early recognition of sepsis using validated severity scores such as SOFA, prompt initiation of source-appropriate antimicrobial therapy, and timely escalation of organ support are likely to remain central to improving survival in this high-risk population. Larger multicentre studies incorporating long-term post-discharge outcomes are warranted to further refine risk stratification and guide resource allocation in critical care settings.