Introduction
Metabolic acidosis is a critical indicator of physiological decompensation in the intensive care unit (ICU), representing a severe departure from homeostatic acid-base equilibrium. Characterized by a primary reduction in serum bicarbonate (HCO-3) and a decrease in arterial pH, it occurs when pathological processes overwhelm pulmonary and renal compensatory mechanisms [1]. In critically ill patients, this condition is rarely an isolated laboratory finding; it serves as a biochemical signal of cellular dysfunction, inadequate tissue perfusion, or systemic organ failure. Whether the onset is acute, as in septic shock, or insidious, as in chronic renal insufficiency, the resulting acidic environment impairs myocardial contractility, disrupts vascular resistance, and precipitates life-threatening electrolyte imbalances like hyperkalemia [2].
The diagnostic approach centers on the serum anion gap (AG), which categorizes acidosis into distinct pathophysiological streams. A high AG typically indicates the accumulation of unmeasured organic acids such as lactate during hypoperfusion or ketoacids in metabolic crises whereas a normal AG signifies direct bicarbonate loss. However, tertiary care ICUs often present multi-factorial complexity, where patients exhibit mixed acid-base disorders and overlapping etiologies, such as sepsis complicated by acute kidney injury (8). Thus, the "clinical profile" of acidosis encompasses the etiology, biochemical severity, and the extent of concomitant organ dysfunction [3].
As regional referral hubs, tertiary care hospitals manage highly complex cases, often characterized by multi-morbidity and advanced disease states. Consequently, the epidemiology of metabolic acidosis in these settings is shaped by a "referral filter," potentially showing a higher prevalence of multi-drug-resistant infections and cases requiring continuous renal replacement therapy (CRRT). Despite its clinical significance, there is a paucity of contemporary data capturing the holistic clinical profile and specific outcomes in our regional tertiary care context [4-5].
This knowledge gap prevents the optimization of evidence-based care, leading to reliance on generalized protocols that may not reflect local clinical realities. Specifically, the prognostic significance of different etiological trends and the efficacy of various therapeutic interventions remain under-researched in this setting. This study aims to systematically analyze the clinical profile and outcomes of metabolic acidosis in a tertiary care ICU. By identifying key prognostic determinants and mapping therapeutic pathways, we seek to provide an evidence base to refine clinical guidelines and improve risk-stratified patient management [5-6].
The present study determined the clinical profile and outcomes of metabolic acidosis in a tertiary care ICU by estimating its frequency, identifying its clinico-etiological patterns, and evaluating their correlation with patient prognosis and survival.
Materials and Methods
Study Design and Setting
This hospital-based, prospective, cross-sectional observational study was conducted within the Intensive Care Unit (ICU) and Emergency Department of Sri Guru Ram Das Institute of Medical Sciences and Research, Amritsar. The study protocol was reviewed and approved by the Institutional Ethics Committee (SGRD/IEC/2024-374), and the research was carried out over an 18-month period from July 2024 to December 2025.
Study Population and Sampling
The study population comprised adult patients (aged≥18 years) admitted to the ICU or emergency department with a primary diagnosis of metabolic acidosis. Present study employed a consecutive sampling method to include all eligible patients during the study period. Inclusion required a documented arterial pH <7.35 with concomitant bicarbonate levels (HCO-3) <22 mEq/L. The pregnant patients and those aged <18 years to maintain a homogenous adult cohort were excluded.
Data Collection and Methodology
Upon enrolment, informed written consent was obtained from the patients or their legal guardians in their vernacular language. A structured proforma was utilized to record detailed demographic profiles, clinical history, comorbidities, and systemic examination findings.
Laboratory Investigations
Arterial blood gas (ABG) analysis was performed at admission (Day 1) and serially for up to five days to monitor biochemical progression. Samples were collected via the radial, brachial, or femoral arteries using heparinized syringes. Complementary laboratory investigations included-Biochemical Profile: Serum electrolytes, renal function tests (RFT), and serum albumin; Metabolic Markers: Serum lactate, blood/urine ketones, and urine pH; Hematological Profile: Complete blood count (CBC).
Acid-Base Classification
The physiological pattern of acidosis was categorized using standard formulas: Anion Gap (AG): Na+ - (Cl- + HCO-3), with high AG defined as >12 mEq/L. Albumin Correction: AG was adjusted by 2.5 mEq/L for every 1 g/dL decrease in serum albumin below 4 g/dL. Delta Ratio: (∆AG / ∆HCO-3) was calculated to identify mixed acid-base disturbances. Ancillary Gaps: Osmolar and urine anion gaps were calculated where indicated to differentiate toxic ingestions or renal versus gastrointestinal bicarbonate losses.
Outcome Measures
Patients were followed throughout their ICU stay to document clinical trajectories. Comparative assessments between Day 1 and Day 5 included hemodynamic stability (heart rate, blood pressure, pO2), requirement for organ support (mechanical ventilation, vasopressors, or renal replacement therapy), and final disposition (discharge or mortality).
Statistical Analysis
Data were analysed using MS-excel software. Continuous variables are presented as mean± standard deviation, and categorical data as frequencies and percentages. The Paired t-test was used for intra-group comparisons between Day 1 and Day 5. Bivariate analysis of survivors versus non-survivors was conducted using the Chi-square or Fisher’s exact test. A p-value <0.05 was considered statistically significant.
Results
The study cohort comprised 109 patients with a mean age of 55.09±16.41 years and a noted male predominance (56.9%). Clinical profiling revealed a high burden of chronic lifestyle diseases, primarily led by hypertension (40.4%) and type 2 diabetes mellitus (39.4%), followed by significant rates of chronic kidney disease (27.5%) and cardiovascular complications (23.9%). Pathophysiological classification of the metabolic derangement showed that Normal Anion Gap Metabolic Acidosis (NAGMA) was the most prevalent subtype, occurring in 57.8% of cases, while High Anion Gap Metabolic Acidosis (HAGMA) was observed in 41.3% of the population. Only a negligible fraction (0.9%) exhibited a mixed acid-base pattern, suggesting that the majority of ICU admissions in this series presented with distinct, single-etiology metabolic acidosis (Table 1).
| Parameter | Value |
| Age (Years), Mean ± SD | 55.09±16.41 |
| Gender (Male: Female) | 62 (56.9%): 47 (43.1%) |
| Primary comorbidities, n(%) | |
| Hypertension | 44 (40.4%) |
| Type 2 Diabetes Mellitus | 43 (39.4%) |
| Chronic Kidney Disease | 30 (27.5%) |
| Cardiovascular Disease | 26 (23.9%) |
| Acidosis Classification, n (%) | |
| NAGMA | 63 (57.8%) |
| HAGMA | 45 (41.3%) |
| Mixed Pattern | 1 (0.9%) |
A longitudinal analysis of arterial blood gas (ABG) parameters demonstrated a highly significant clinical recovery in the cohort’s acid-base status over the first five days of intensive care. The mean pH improved from a baseline of 7.203±0.143 to 7.336±0.140 by Day 5 (p< 0.001), effectively shifting the average clinical state from significant acidemia toward the normal physiological range. This was paralleled by a substantial rise in serum bicarbonate levels, which increased from 15.71±10.87 mEq/L to 20.23±10.27 mEq/L (p<0.001), indicating successful correction of the underlying metabolic deficit (Table 2). While mean lactate levels showed a downward trend from 2.908 mmol/L to 2.449 mmol/L, this change did not reach statistical significance (p=0.246), suggesting that while systemic acidosis was largely corrected, tissue-level metabolic stress remained variable across the study population. Similarly, the stability of PaCO2 levels (p=0.739) indicates that the observed pH recovery was primarily driven by metabolic stabilization rather than respiratory compensation (Table 2).
| Parameter | Day 1 (Mean± SD) | Day 5 (Mean ± SD) | p-value* |
| pH | 7.203± 0.143 | 7.336±0.140 | <0.001 |
| Bicarbonate (HCO-3) (mEq/L) | 15.71±10.87 | 20.23 ±10.27 | <0.001 |
| Lactate (mmol/L) | 2.908±2.765 | 2.449±3.444 | 0.246 |
| PaCO2 (mmHg) | 37.25±26.20 | 38.20±22.31 | 0.739 |
Comparative analysis between survivors and non-survivors reveals several critical physiological determinants of mortality in patients with metabolic acidosis (Table 3). Non-survivors exhibited significantly lower systolic blood pressure (96.25±15.98 mmHg vs. 117.05±20.88 mmHg; p=0.007) and more profound admission acidemia (pH 7.109±0.159 vs. 7.214±0.135; p=0.009), underscoring the lethal synergy of hemodynamic instability and severe metabolic derangement. Impaired oxygenation and hepatic dysfunction also emerged as vital prognostic markers, with non-survivors showing significantly lower PaO2 levels (p=0.020) and markedly elevated total bilirubin (p=0.009). Notably, the clinical trajectory by Day 5 served as a definitive predictor of outcome; non-survivors had a significantly higher requirement for inotropic support and a greater proportion of persistent clinical decompensation (p<0.001 Table 3). These findings suggest that early hemodynamic resuscitation and the reversal of metabolic parameters within the initial phase of care are paramount for survival.
| Parameter | Survivors (n=100) | Non-survivors (n=9) | p-value |
| Systolic BP (mmHg) | 117.05±20.88 | 96.25±15.98 | 0.007 |
| Admission pH | 7.214±0.135 | 7.109±0.159 | 0.009 |
| PaO2 (Day 1) (mmHg) | 104.60±57.51 | 83.17±42.89 | 0.020 |
| Total Bilirubin (mg/dL) | 1.66±5.07 | 88.65±260.01 | 0.009 |
| Inotrope Use (Day 5), n (%) | 7 (43.8%) | 9(56.3%) | <0.001 |
| Decompensated (Day 5), n (%) | 27 (73.0%) | 10 (27.0%) | <0.001 |
The longitudinal assessment of therapeutic interventions highlights a progressive stabilization of the patient cohort over the five-day observation period (Table 4). While the requirement for invasive ventilation and renal replacement therapy remained relatively substantial, decreasing marginally from 39.4% to 33.9% and 28.4% to 24.8% respectively, a more pronounced reduction was observed in the necessity for inotropic support, which halved from 28.4% on admission to 14.7% by Day 5. This trend reflects a successful mitigation of acute circulatory failure in a significant subset of the population. Ultimately, the intensive management strategy yielded a cumulative survival rate of 91.7%, with an overall in-hospital mortality of 8.3%, underscoring the critical nature of sustained multi-organ support in the management of severe metabolic acidosis.
| Intervention | Day 1, n (%) | Day 5, n (%) |
| Invasive Ventilation | 43 (39.4%) | 37 (33.9%) |
| Inotropic Support | 31 (28.4%) | 16 (14.7%) |
| Renal Replacement Therapy | 31 (28.4%) | 27 (24.8%) |
| Survival Status | Survived: 100 (91.7%) | Mortality: 9 (8.3%) |
Discussion
The present study observed an overall in-hospital mortality rate of 8.3%, which is significantly lower than the figures reported in recent literature concerning extreme acidosis. Previous investigations by Henrique et al.,[7] and Allyn et al.,[8] reported mortality rates of 42% and nearly 70%, respectively. This disparity likely stems from the inclusion criteria; while the comparative literature focused on extreme acidosis (pH < 7.0), our cohort included a broader spectrum (pH<7.35). Furthermore, our high survival rate of 91.7% correlates with the observation that metabolic etiologies, particularly those related to Type 2 Diabetes Mellitus (39.4%), typically carry a more favorable prognosis than etiologies like cardiac arrest. As noted in the comparative text, diabetic-related pathologies (such as DKA) often involve beneficial physiological effects from ketone production [9,10].
A major point of discussion in critical care is whether the depth of acidemia or the primary diagnosis is the superior predictor of death. Henrique et al.,[7] hypothesized that at very low pH levels, the acidosis itself might exert a greater influence on outcomes than the primary disorder. However, our results align more closely with the alternative view that the underlying diagnosis is paramount. While our non-survivors had significantly lower admission pH (7.109±0.159) compared to survivors (7.214±0.135; p=0.009 Table 3), the comparative literature emphasizes that pH alone, without considering etiology, should not justify the cessation of treatment. This is evidenced by high survival in conditions like seizures, which can present with extreme, albeit transient, acidosis (pH < 6.8) [11,12].
The comparative literature indicates that it typically requires approximately 48 hours for a patient’s pH to return to physiological norms. Our longitudinal data supports this timeframe, as we observed a highly significant recovery from a baseline mean pH of 7.203 to 7.336 by Day 5 (p<0.001). Interestingly, while our systemic pH and bicarbonate (15.71 to 20.23 mEq/L, Table 2) corrected significantly, lactate levels did not reach statistical significance in their downward trend (p=0.246). This suggests a recovery lag where biochemical normalization of the blood precedes the resolution of tissue-level metabolic stress, a trajectory similar to that seen in patients following out-of-hospital cardiac arrest [13].
Beyond acid-base parameters, hemodynamic stability emerged as a critical determinant of survival in our study. Non-survivors exhibited significantly lower systolic blood pressure (96.25±15.98 mmHg vs. 117.05±20.88 mmHg; p=0.007. Table 3) and a markedly higher requirement for inotropic support by Day 5 (p<0.001). This matches the comparative findings where progressive acidosis—often seen in cardiac arrest or prolonged ischemia—leads to worsening pathology, such as myocardial dysfunction [14,15]. The persistence of inotropic needs and clinical decompensation by the fifth day served as definitive predictors of a poor outcome in our population.
This study reinforces the importance of considering both the type and the etiology of acidosis when determining prognosis. Our findings suggest that once acid-base derangement is present, absolute pH values may be less useful than other markers like age, lactate, or hemodynamic status. Furthermore, the fact that a significant portion of patients in some studies (20%) may not require ICU admission but still face high 30-day mortality (40%) highlights the necessity for rigorous follow-up and active treatment, regardless of the initial site of care. In our setting, early aggressive resuscitation and the successful mitigation of circulatory failure evidenced by the halving of inotropic requirements by Day 5 were central to achieving a high cumulative survival rate.
Limitations
Sample Size and Scope: The study was conducted at a single center with a cohort of 109 patients, which may limit the generalizability of the findings to different geographical or resource-limited settings.
Conclusion
This study underscores that metabolic acidosis in the tertiary ICU setting is primarily a manifestation of chronic lifestyle-related organ dysfunction and acute physiological stress. While standard intensive care protocols effectively corrected the biochemical pH and bicarbonate levels in over 90% of the cohort, the metabolic recovery did not always equate to tissue-level recovery as evidenced by non-significant lactate trends. Admission pH, systolic hypotension, and the persistence of inotropic requirement by the fifth day serve as the most reliable indicators of a poor prognosis.
Future Recommendations
Protocol Refinement: Future research should focus on developing a Metabolic Risk Score that integrates admission pH with lactate clearance rates to trigger earlier aggressive intervention. Expanding this research to a multicentric level would help map regional etiological variations (e.g., toxic ingestions vs. septic NAGMA) across different demographic strata. Future longitudinal studies should track the progression of chronic kidney disease in survivors of severe ICU-acquired metabolic acidosis to assess the permanent physiological "cost" of these acute episodes.
Abbreviations
ABG: Arterial Blood GasAG: Anion GapCBC: Complete Blood CountCKD: Chronic Kidney DiseaseCRRT: Continuous Renal Replacement TherapyHAGMA: High Anion Gap Metabolic AcidosisHCO₃⁻: BicarbonateNAGMA: Normal Anion Gap Metabolic AcidosisPaCO₂: Partial Pressure of Carbon DioxidePaO₂: Partial Pressure of OxygenRFT: Renal Function Test
Declarations
Ethical Approval and Consent to Participate
The study protocol was reviewed and approved by the Institutional Ethics Committee of Sri Guru Ram Das Institute of Medical Sciences and Research, Sri Guru Ram Das University of Health Sciences, Amritsar, Punjab, India (Approval No. SGRD/IEC/2024-374). The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. Written informed consent was obtained from all participants or their legally authorized representatives before enrolment in the study.
Consent for Publication
Not applicable. The manuscript does not contain any individual person's identifiable information, photographs, or personal data requiring consent for publication.
Availability of Supporting Data
The datasets generated and/or analysed during the current study are available from the corresponding author upon reasonable request.
Competing Interests
The authors declare that they have no competing interests.
Funding
This research received no external funding. The study was conducted using the institutional facilities and departmental resources of Sri Guru Ram Das Institute of Medical Sciences and Research, Amritsar.
Authors' Contributions
Aasra Kaur Brar: Conceptualization, patient recruitment, data collection, clinical assessment, literature review, and preparation of the initial manuscript draft.
Gurinder Mohan: Study conception and design, supervision of the study, clinical oversight, interpretation of data, critical revision of the manuscript, and final approval.
Manish Chandey: Clinical supervision, interpretation of findings, manuscript review, and critical intellectual input.
Amitpal Kaur: Data acquisition, statistical interpretation, literature review, and manuscript editing.
Prabhjot Kaur Gill: Statistical analysis, interpretation of results, manuscript editing, language revision, and final approval of the manuscript.