Sepsis is a complex syndrome resulting from a dysregulated host response to infection, leading to life-threatening organ dysfunction and high risk of death. According to the Sepsis-3 definitions, the presence of infection plus organ dysfunction, as quantified by an increase in the Sequential Organ Failure Assessment (SOFA) score, defines sepsis, and when accompanied by circulatory and cellular/metabolic abnormalities, defines septic shock. Early recognition of severity and stratification of risk are essential in guiding therapy, allocating resources, and improving outcomes. However, clinical signs alone may be insufficiently specific, and laboratory biomarkers are employed to support prognostication. A major pathophysiologic feature of sepsis is tissue hypoperfusion, which results in anaerobic metabolism and accumulation of lactate in the bloodstream. Elevated serum lactate concentration reflects not only diminished perfusion but also increased metabolic demand, mitochondrial dysfunction, and impaired clearance (notably by liver). Thus, lactate is a candidate biomarker for assessing severity, guiding resuscitation, and predicting mortality. Serial measurements (admission, 6 hours, 24 hours, etc.) allow evaluation of lactate kinetics or clearance, which may offer better prognostic value than a single measurement. Several large studies and meta-analyses have demonstrated the association between elevated blood lactate and increased mortality in sepsis. A network meta-analysis including 127 studies and over 107,000 patients found that higher blood lactate is significantly associated with sepsis mortality (odds ratio ~1.57) and that non-survivors have meaningfully higher lactate than survivors. It also showed that lactate has a reasonably good predictive accuracy for mortality (area under ROC ~0.72). [1] Comparisons of lactate vs lactate clearance have also been done: in patients with septic shock by Sepsis-3 criteria, one study found that both initial lactate and 6-hour lactate clearance are associated with 28-day mortality, but the lactate level at 6 hours had superior prognostic value (AUC ~0.70 vs ~0.65) than clearance alone. [2] Cut-off values for lactate vary between studies. A retrospective cohort study of severe sepsis/septic shock patients found that an initial lactate > 2.5 mmol/L was the best threshold to predict 28-day mortality (with sensitivity ~67%, specificity ~62%) in their cohort. [3] Another study in adult sepsis/septic shock with initial hyperlactatemia (lactate ≥2 mmol/L) showed that non-survivors had higher 6-hour lactate levels and lower 6-hour lactate clearance; metrics such as clearance <10–30% were significant predictors. [4] Pediatric studies also support the value of lactate and its kinetics. In a prospective pediatric septic shock cohort, admission lactate ≥4 mmol/L was a strong predictor of mortality; a 24-hour lactate clearance <10% carried a high predictive value for death. [5] In children with sepsis (not necessarily shock), a study using a large database found that for every 1 mmol/L increase in lactate, there was roughly a 17% increased hazard of 28-day mortality, with a threshold around 2.2 mmol/L distinguishing higher risk. [6] Studies in low- and middle-income countries, including India, have explored lactate in local settings. One prospective Indian study in elderly septic patients (>60 years) showed that mean serum lactate in non-survivors was significantly higher than in survivors, and that lactate correlated with severity (sepsis → severe sepsis → septic shock). [7] Another recent study in pregnancy-associated sepsis in India showed that admission lactic acid and serial levels were significantly higher in those requiring ICU care vs non-ICU, and that a fall (clearance) of more than ~60% was linked with 100% survival, while an increase of 100% predicted universal mortality in their cohort. [8] Despite these findings, several gaps remain. The precise timing for lactate measurement (which intervals provide maximum discriminatory power), the optimal thresholds in different populations, and how lactate and lactate clearance compare to or complement scoring systems like SOFA, APACHE II or qSOFA are not uniformly established. Moreover, resource constraints in settings like many Indian hospitals, including Sher-e-Kashmir Institute of Medical Sciences (SKIMS), Soura, may impact feasibility of frequent lactate measurements or serial assessments. Local data are sparse, particularly in the Kashmiri population, which may have different baseline comorbidities, healthcare-access delays, and patterns of sepsis. Therefore, this prospective study at Sher-e-Kashmir Institute of Medical Sciences (SKIMS), Souraaims to assess admission serum lactate levels and serial lactate measurements (e.g., at 6 h, 24 h) in adult sepsis patients, evaluate lactate clearance, and analyze associations with mortality, ICU stay duration, requirement for vasopressors and mechanical ventilation. The study also aims to derive locally relevant cut-off values and compare the prognostic accuracy of lactate metrics with standard clinical predictors.

Study design and setting This was a prospective observational study conducted in the Department of Medicine and Intensive Care Unit (ICU) at Sher-e-Kashmir Institute of Medical Sciences (SKIMS), Soura over a period of 18 months, from January 2023 to June 2024. The hospital is a tertiary care referral center in Jammu and Kashmir, catering to a large population with diverse socio-demographic and clinical profiles. Study population All adult patients aged 18 years and above admitted to the ICU with a diagnosis of sepsis according to the Sepsis-3 definition were considered eligible. Sepsis was defined as suspected or documented infection with an increase in Sequential Organ Failure Assessment (SOFA) score of ≥2 points. Patients with known inborn errors of metabolism, malignancy with terminal illness, advanced chronic liver disease (Child-Pugh C), or those who declined consent were excluded from the study. Sample size A total of 120 patients were enrolled consecutively during the study period. Sample size estimation was based on previous observational studies showing mortality rates of 30–50% in septic patients with elevated lactate levels [1]. Assuming a power of 80% and a significance level of 5%, a minimum of 100 patients were required, and 120 patients were included to account for possible dropouts. Data collection Demographic information (age, sex, residence), comorbidities (diabetes, hypertension, chronic kidney disease, chronic lung disease), suspected source of infection, and clinical parameters at admission were recorded. The SOFA score and quick SOFA (qSOFA) score were calculated at baseline. Lactate measurement Venous blood samples were collected for serum lactate estimation at three time points: at admission (0 hours), at 6 hours, and at 24 hours. Lactate was measured using an enzymatic colorimetric method with a standardized laboratory analyzer (Roche Cobas c311, Mannheim, Germany). Lactate clearance was calculated using the formula: Lactate clearance (%) = \[(Lactate at 0 h – Lactate at 6 or 24 h) ÷ Lactate at 0 h] × 100. Patients were categorized into groups based on admission lactate (<2 mmol/L, 2–4 mmol/L, and >4 mmol/L) and lactate clearance (<10%, 10–30%, >30%). Outcome measures Patients were followed until hospital discharge or death. Primary outcome was in-hospital mortality. Secondary outcomes included duration of ICU stay, need for vasopressors, requirement for mechanical ventilation, and length of hospital stay. Statistical analysis Data were analyzed using Statistical Package for Social Sciences (SPSS) version 25.0 (IBM Corp., Armonk, NY, USA). Continuous variables were expressed as mean ± standard deviation (SD) or median (interquartile range) as appropriate, and categorical variables as frequencies and percentages. Student’s t test or Mann–Whitney U test was applied for continuous variables, and chi-square or Fisher’s exact test for categorical variables. Logistic regression analysis was used to determine independent predictors of mortality. A p value <0.05 was considered statistically significant.

A total of 120 patients with sepsis were enrolled in the study. The mean age was 54.7 ± 15.3 years, with a range of 18–85 years. Males constituted 58% of the study population. The most common comorbidities were hypertension (32.5%) and diabetes mellitus (30%). The predominant sources of infection were respiratory tract (34%), urinary tract (22%), intra-abdominal infections (18%), and skin/soft tissue infections (10%). The baseline demographic and clinical characteristics of the patients are shown in Table 1. Table 1. Demographic and baseline clinical characteristics of the study population (n = 120) Variable Value Mean age (years) 54.7 ± 15.3 Age group <40 years 22 (18.3%) Age group 40–60 years 49 (40.8%) Age group >60 years 49 (40.8%) Sex (male/female) 70 (58.3%) / 50 (41.7%) Hypertension 39 (32.5%) Diabetes mellitus 36 (30.0%) Chronic kidney disease 14 (11.7%) Chronic lung disease 10 (8.3%) Source of infection – respiratory 41 (34.2%) Source of infection – urinary tract 26 (21.7%) Source of infection – intra-abdominal 22 (18.3%) Source of infection – skin/soft tissue 12 (10.0%) Source of infection – others (meningitis, endocarditis, etc.) 19 (15.8%) Mean SOFA score at admission 7.4 ± 2.9 Patients were stratified into three groups based on admission lactate levels: <2 mmol/L, 2–4 mmol/L, and >4 mmol/L. Nearly 42% of patients had admission lactate >4 mmol/L, while 28% were between 2 and 4 mmol/L, and 30% had lactate <2 mmol/L. The distribution is shown in Table 2. Table 2. Distribution of patients according to admission serum lactate levels Lactatecategory Number of patients (%) <2 mmol/L 36 (30.0%) 2–4 mmol/L 34 (28.3%) >4 mmol/L 50 (41.7%) Outcomes including mortality, ICU stay, requirement for vasopressors, and need for mechanical ventilation were compared among the three lactate groups. Patients with lactate >4 mmol/L had significantly higher mortality (58%) compared to those with lactate 2–4 mmol/L (32%) and <2 mmol/L (22%). They also required more vasopressors and had longer ICU stays. These findings are summarized in Table 3. Table 3. Outcomes according to admission serum lactate levels Outcome Lactate <2 mmol/L (n=36) Lactate 2–4 mmol/L (n=34) Lactate >4 mmol/L (n=50) P value Mortality (%) 8 (22.2%) 11 (32.4%) 29 (58.0%) <0.01 Mean ICU stay (days) 6.8 ± 2.3 8.2 ± 3.1 11.4 ± 4.0 <0.01 Vasopressor requirement 12 (33.3%) 18 (52.9%) 38 (76.0%) <0.01 Mechanical ventilation 10 (27.8%) 14 (41.2%) 32 (64.0%) <0.01 Serial lactate measurements were analyzed at 6 hours and 24 hours. Survivors demonstrated significant lactate clearance at both intervals, whereas non-survivors showed persistently elevated or rising lactate levels. At 6 hours, mean lactate clearance in survivors was 21.4%, compared to 5.8% in non-survivors. At 24 hours, clearance reached 39.2% in survivors but only 9.6% in non-survivors. Details are shown in Table 4. Table 4. Lactate clearance at 6 and 24 hours in survivors vs non-survivors Variable Survivors (n=72) Non-survivors (n=48) P value Admission lactate (mmol/L) 3.1 ± 1.6 4.7 ± 2.2 <0.01 6-hour lactate (mmol/L) 2.4 ± 1.2 4.5 ± 2.1 <0.01 24-hour lactate (mmol/L) 1.9 ± 0.9 4.3 ± 2.0 <0.01 6-hour clearance (%) 21.4 ± 12.6 5.8 ± 8.4 <0.01 24-hour clearance (%) 39.2 ± 18.1 9.6 ± 10.3 <0.01 Multivariate logistic regression was performed to identify independent predictors of in-hospital mortality. After adjusting for age, comorbidities, and SOFA score, admission lactate ≥4 mmol/L and poor lactate clearance (<10% at 24 hours) were independent predictors of mortality. Results of the regression model are shown in Table 5. Table 5. Multivariate logistic regression analysis of predictors of mortality Predictor Odds ratio (OR) 95% confidence interval (CI) P value Age >60 years 1.8 1.0–3.4 0.05 Diabetes mellitus 1.5 0.8–2.9 0.21 SOFA score ≥8 2.7 1.4–5.2 <0.01 Admission lactate ≥4 mmol/L 3.5 1.8–6.9 <0.01 Lactate clearance <10% at 24 h 4.1 2.0–8.4 <0.01

In our study of 120 adult sepsis patients, an admission serum lactate ≥4 mmol/L was associated with significantly higher in-hospital mortality, longer ICU stays, and increased need for vasopressors and mechanical ventilation. Similarly, patients who survived showed much greater lactate clearance at 6 and 24 hours compared to non-survivors, and multivariate analysis confirmed that admission lactate ≥4 mmol/L and less than 10% clearance at 24 hours were independent predictors of mortality. These findings are consistent with pediatric data from Choudhary et al. In a study of children with septic shock in North India: nonsurvivors had significantly higher admission lactate (mean ~5.12 vs ~3.13 mmol/L in survivors) and lactate clearance <10% at 24 hours was a strong predictor of mortality (sensitivity ~78.7%, specificity ~72.2%) [9]. That matches our clearance threshold and prognostic value. A systematic review and meta-analysis by Zhang, Xu, and Ni et al. Demonstrated that lactate clearance is strongly predictive of mortality in critically ill patients. The pooled sensitivity was ~0.75 and specificity ~0.72, which align with our observations that higher clearance predicts survival and low clearance is risky [10]. The retrospective cohort from the Netherlands (Baysan et al.) on adult ICU sepsis patients showed that including lactate at 24 hours (alone or with admission lactate / 24-hour clearance) in mortality prediction models (e.g., APACHE IV) improves their discrimination significantly. This supports our result that the dynamic measure (clearance) gives additional prognostic information beyond the admission value [11]. Also, pediatric work by study “Serum lactate levels as the predictor of outcome in pediatric septic shock” (30 children) found that lactate values >5 mmol/L at admission, 12h, and 24h were significant predictors of death, with increasing odds ratios at each time point (e.g. OR ~12.5 at 12 h) [12]. Though their threshold is slightly higher than ours, the pattern—higher lactate correlating with mortality—is the same. Our study’s observation that patients with admission lactate >4 mmol/L had \~58% mortality is somewhat higher than in many adult studies, but similar to high-mortality paediatric septic shock cohorts. Possible reasons include late presentation, limited resources, or more severe illness on admission. Consistent with our findings, the meta-analysis also indicated that better clearance (over 6-24 h) is protective; non-survivors had poor clearance or even worsening values. This suggests that serial monitoring is essential. It is not enough to measure lactate once; how it changes over time gives better prognostic insight. Clinical implications of our results include: using an admission lactate threshold of \~4 mmol/L to trigger more aggressive therapy; using lactate clearance at 24 hours as part of sepsis management protocols to assess response; possibly combining static lactate with severity scores (like SOFA or APACHE) to improve prognostic accuracy; resource-allocation decisions in ICUs (e.g. closer monitoring, earlier mechanical ventilation, or vasopressor use) for patients with high lactate and low clearance. Limitations: Our sample is moderate, which may limit precision of estimates. There could be selection bias (patients admitted to ICU represent more severe cases). Our study did not measure pre‐ICU lactate levels or interventions before admission, which might affect baseline lactate. Also, comorbidities (liver dysfunction etc.) or non-sepsis causes of elevated lactate were not fully stratified. Finally, external validation in other centers is needed to generalize our findings beyondSher-e-Kashmir Institute of Medical Sciences (SKIMS) . In sum, our results support and extend existing literature: admission lactate ≥4 mmol/L and poor clearance (especially <10% at 24 h) are strong predictors of mortality in sepsis. Combined static and dynamic lactate measurements should be integrated into sepsis protocols in settings similar to ours.

This prospective study conducted at Sher-e-Kashmir Institute of Medical Sciences (SKIMS), Soura demonstrates that both admission serum lactate levels and lactate clearance are powerful prognostic markers in patients with sepsis. Patients presenting with an initial lactate ≥4 mmol/L had significantly higher in-hospital mortality, longer ICU stays, and increased need for mechanical ventilation and vasopressors. Furthermore, inadequate lactate clearance, particularly less than 10% at 24 hours, independently predicted poor outcomes regardless of baseline severity scores. These findings highlight the importance of incorporating both static and dynamic lactate measurements into sepsis management protocols. Admission lactate provides early risk stratification, while serial monitoring and calculation of clearance offer valuable insight into the effectiveness of resuscitative efforts and evolving patient prognosis. In resource-limited settings such as ours, even two measurements (at admission and 24 hours) can significantly aid decision-making. Future research with larger, multicenter cohorts is required to validate these results and to assess whether lactate-guided therapeutic strategies can improve survival. Until then, routine lactate monitoring should be considered a vital component of sepsis care in ICUs. Conflict of interest: Nil Funding: Nil

1. Zhang L., Hu G., Gu L., Kan L., Xia Z., Dong W., Cai X., Xu Y., Yuan Z., 2023. "Hierarchical Capability in Distinguishing Severities of Sepsis via Serum Lactate: A Network Meta-Analysis", Critical Care, 27(1):109. 2. van der Slikke J.W., Holleman F., Fügel D., Martin-Loeches I., Leone M., Moerbeek M., Peetermans W., Wiersinga W\.J., Schouten J., 2018. "Lactate Level Versus Lactate Clearance for Predicting Mortality in Patients With Septic Shock Defined by Sepsis-3", Critical Care, 22(1):112. 3. Silversides J.A., Major E., Ferguson A., Day A., Lammi M., Ball I., Morrow L., 2016. "Blood Lactate Levels Cutoff and Mortality Prediction in Sepsis-Time for a Reappraisal?", Journal of Intensive Care Medicine, 31(7):493-498. 4. Lee S.G., Song J., Park D.W., Moon S., Cho H.-J., Kim J.Y., Park J., Cha J.H., 2021. "Prognostic value of lactate levels and lactate clearance in sepsis and septic shock with initial hyperlactatemia: A retrospective cohort study according to the Sepsis-3 definitions", Medicine (Baltimore), 100(7)\:e24835. 5. Kumar P., Kulkarni M., de Araujo M.P., Pollack M.M., Michelson N., 2017. "Lactate clearance as the predictor of outcome in pediatric septic shock", Pediatric Critical Care Medicine, 18(11):1065-1071. 6. Song Y., Wang N., Xie X., Tian Y., Wang Y., 2024. "Relationship between lactate levels and 28-day mortality in pediatric sepsis: results from the pediatric intensive care database", BMC Pediatrics, 24:712. 7. Anand A., Gambhir I.S., Nath G., Kumar N., 2013. "Serum Lactate as Prognostic Marker in Indian Elderly with Septicemia", National Journal of Medical Research, 3(4):350-352. 8. Singh A., Bhat S., Dubey S., Kaur P., Jain B., 2020. "Serial Serum Lactic Acid in Pregnancy-Associated Sepsis for Maternal Outcome", Journal of Obstetrics and Gynaecology India, 70(6):457-463. 9. 9.Choudhary R., Sitaraman S., Choudhary A., 2017. “Lactate clearance as the predictor of outcome in pediatric septic shock”, Journal of Emergencies, Trauma and Shock, 10(2):55-59. 10. Zhang Z., Xu X., Ni H., 2014. “Lactate clearance is a useful biomarker for the prediction of all-cause mortality in critically ill patients: a systematic review and meta-analysis”, Critical Care Medicine, 42(9):2149-2159. 11. Baysan M., Baroni G.D., van Boekel A.M., Steyerberg E.W., Arbous M.S., van der Bom J.G., 2020. “The added value of lactate and lactate clearance in prediction of in-hospital mortality in critically ill patients with sepsis”, Critical Care Explorations, 2(3)\:e0087. ([PubMed][3]) 12. Natraj S., Subramanian S., Mani K., 2011. “Serum lactate levels as the predictor of outcome in pediatric septic shock”, Indian Pediatrics, 48(8):599-604.