Clinical Predictors of Severe and Fatal COVID-19: Inflammatory and Haematological Biomarkers in Early Risk Stratification

INTRODUCTION

SARS-CoV-2 is responsible for COVID-19, which began in Wuhan, China, in December 2019.^[1] The World Health Organization declared it a global pandemic in March 2020.^[2] Healthcare systems were under significant pressure during the pandemic. COVID-19 is highly contagious, with 81% experiencing mild illness, 14% developing severe disease, and 5% progressing to critical illness.^[3,4] Severe cases can lead to complications like respiratory distress and coagulopathy, with a high mortality rate associated with critical disease. Biomarkers like interleukin-6 (IL-6), D-dimer, lymphocyte count (LC), neutrophil count (NC), white blood corpuscles (WBC) count, neutrophil-to-lymphocyte ratio (NLR), activated partial thromboplastin time (aPTT), and prothrombin time (PT) are linked to immune-inflammatory and coagulation pathways. Other biomarkers like transaminases and lactate dehydrogenase (LDH) indicate cellular inflammation and damage and are used to assess disease severity.^[5,6] However, only a few of these biomarkers can reliably predict therapy responses or indicate poor outcomes in patients with COVID-19.^[7] Inflammatory markers are released as a natural immune response to virus infection; however, they can be harmful if released in excess, they include D-dimer, C-reactive protein (CRP), ferritin, neutrophils, platelets, and LDH.^[7,8] Higher levels of these markers typically correlate with more severe infections and can indicate a poor prognosis, outcome, and higher mortality risk.^[9,10] While elevated inflammatory markers are also seen in other conditions, they are useful in identifying and assessing the extent of COVID-19, predicting complications like acute respiratory distress syndrome (ARDS), and monitoring treatment responses. A reduction in markers such as CRP can signal that treatment is effective.^[11]

According to several studies, COVID-19 can lead to coagulopathy, resulting in disseminated intravascular coagulopathy (DIC). Key indicators of coagulopathy include high D-dimer and ferritin levels, while anomalies in platelet count, PT, and PTT are less common. Early elevation of inflammatory mediators like CRP, LDH, ferritin, and IL-6, along with coagulation tests,^[12] is essential for detecting COVID-19.^[7,8,13-15] A systematic review by Zaki et al.^[16] found that comorbidities like hypertension, diabetes, stroke, ischemic events, cancer, renal diseases, and hypercholesterolemia are closely linked to patients with COVID-19 outcomes. Diabetes is linked to worse outcomes, while hypertension is a significant predictor of severe disease. Pre-established kidney conditions are more likely to experience acute kidney injury from COVID-19.^[16] A study by Oussalah et al.^[17] found that kidney function indicators, particularly high urea nitrogen levels, were strong predictors of death in severe COVID-19. Levels of inflammatory markers, lymphopenia, and high NCs were associated with critical illness in a study by Tjendra et al.^[9] Serious consequences, including systemic vasculitis and organ failure, might be indicated by a D-dimer, as addressed by Ponti et al.^[10]

Theoretically, COVID-19 may affect various organs systemically because it targets endothelium, according to Sardu et al.^[18] Evidence suggests that hypertension, clotting disorders, and kidney disease are all influenced by endothelial cell failure. Endothelial cells amplify the expression of angiotensin-converting enzyme 2 (ACE2). COVID-19’s effect on endothelial cells has the potential to affect outcomes by aggravating underlying diseases like diabetes and hypertension.^[18] Rus et al.^[19] emphasised that COVID-19 exacerbated cardiovascular complications, increasing the risk of AMI and poor prolonged consequences. Sritharan et al.^[20] found that patients with CVD had a higher hospitalisation death rate and were more likely to suffer myocardial injuries. Similarly, Vu et al.^[21] found that myocardial injury in patients with COVID-19 increased in-hospital mortality. Kaftan et al.^[22] evaluated the diagnostic reliability of CRP, ferritin, LDH, and D-dimer in COVID-19 cases in Iraq, and they found that combinations of these markers showed varying diagnostic performance for COVID-19. The main aim in the current study was to assess the effect of demographic factors, comorbidities, and laboratory biomarkers on disease progression, severity, and mortality associated with COVID-19. We aimed to identify the diagnostic and prognostic performance of the studied biomarkers that were monitored longitudinally to study their impact on the disease progression, severity, worse outcome, and mortality.

MATERIAL AND METHODS

RESULTS

Impact of comorbidities on COVID-19 mortality rates

Individuals with comorbidities, such as hypertension, diabetes, asthma, CKD, CVD, and lung diseases, are at an increased risk of experiencing severe COVID-19 outcomes and death. This heightened risk is largely because comorbid conditions can weaken the immune system, making it harder for the body to combat the virus. Additionally, some comorbidities can cause damage to the lungs and other organs, further increasing vulnerability to COVID-19 infection. Figure 1 shows that the mortality rate for COVID-19 infection is significantly higher in individuals with hypertension (p <0.0001), diabetes (p <0.001), and asthma (p <0.0001), compared to those who recovered. However, no significant difference in mortality rates was observed between individuals with CKD, CVD, or lung diseases and those who recovered (p >0.05). On the other hand, the severe cases rate for COVID-19 infection was significantly (p ≤0.001) higher in individuals with hypertension, while non-significant in those with diabetes, asthma, CKD, and CVD compared to non-severe cases.

Close

Figure 1: Prevalence of co-morbidities in total, survived, and deceased groups with corresponding p values. p ≤0.05 is significant. CKD: Chronic kidney disease, CVD: Cardiovascular disease.

Export to PPT

Laboratory inflammatory biomarkers

The analysis of D-dimer levels in COVID-19 cases revealed a statistically significant association with both disease severity (severe vs. non-severe) and patient outcomes. The mean D-dimer levels in deceased patients (4.3 ± 1.5 μg/mL) were significantly higher (p <0.0001) compared to those in recovered patients (1.7 ± 0.8 μg/mL). Similarly, a significant difference in D-dimer levels was observed between severe cases (2.6 ± 1.2 μg/mL) and non-severe cases (0.8 ± 0.4 μg/mL), with a p <0.001. These findings suggest that elevated D-dimer levels may serve as a potential biomarker for both disease severity and mortality risk in patients with COVID-19, as illustrated in Figure 2.

Close

Figure 2: The box-and-whisker plot illustrates the levels of three biomarkers, D-dimer, CRP, and ferritin, across four patient conditions: survived, deceased, non-severe, and severe. For each biomarker, the box plots highlight the variability and distribution within each condition, with the median and interquartile range shown. Generally, higher biomarker levels are observed in deceased and severe patients compared to survived and non-severe groups, indicating a correlation between elevated biomarker levels and more severe outcomes. Specifically, D-dimer and CRP levels are markedly higher in the deceased and severe groups, suggesting their potential utility as indicators of severity. p ≤0.05 is significant. CRP: C-reactive protein.

Export to PPT

Additionally, the analysis revealed a statistically significant difference in the mean CRP levels between survived and deceased patients with COVID-19 (p <0.0001). Recovered patients exhibited lower mean CRP levels (101 ± 42 mg/dL) compared to deceased patients (189 ± 58 mg/dL). A similar significant difference in CRP levels was found between severe and non-severe cases (p <0.001), with severe cases having higher mean CRP levels (132.3 ± 61.3 mg/dL) compared to non-severe cases (77.5 ± 37.8 mg/dL). These findings suggest that CRP levels may be a potential biomarker for both disease severity and prognosis in patients with COVID-19, as shown in Figure 2.

In addition, ferritin, a protein that stores iron, plays a complex role during infection. While it facilitates immune responses, excessively high levels can become detrimental. This study investigated the association between serum ferritin levels and COVID-19 severity and mortality. Patients who succumbed to COVID-19 exhibited significantly higher mean ferritin levels (478 ± 115 ng/mL) compared to those who recovered (288.7 ± 95.87 ng/mL), with a p value <0.001. Similarly, patients with severe COVID-19 had elevated ferritin levels (367.4 ± 129.6 ng/mL) compared to those with non-severe cases (156.9 ± 67.4 ng/mL), with a p value <0.0001. These findings, presented in Figure 2, suggest a strong positive correlation between elevated ferritin levels and both disease severity and mortality. The results highlight the potential of ferritin as a biomarker for predicting COVID-19 severity and mortality risk. Identifying patients with elevated ferritin levels could enable earlier intervention and potentially improve patient outcomes.

Laboratory complete blood count profile

The hematological parameters show significant differences between severe and non-severe cases and between deceased and survived patients. HB levels were lower in severe cases (10.71 ± 1.35 g/dL) compared to non-severe cases (11.82 ± 1.16 g/dL, p <0.001), and in deceased cases (10.10 ± 1.5 g/dL) compared to survivors (11.20 ± 1.9 g/dL, p = 0.06). Similarly, red blood corpuscles (RBC) count was reduced in severe cases (4.65 ± 0.23 ×10⁶/μL) versus non-severe cases (4.98 ± 0.29 ×10⁶/μL, p <0.001) and in deceased cases (4.45 ± 0.70 ×10⁶/μL) compared to survivors (4.60 ± 0.75 ×10⁶/μL, p = 0.08). Hematocrit followed the same trend, being lower in severe (38.8 ± 3.4%) than non-severe cases (40.9 ± 4.2%, p <0.001) and in deceased (39.5 ± 2.7%) compared to survivors (40.0 ± 2.3%, p = 0.06) [Figure 3].

Close

Figure 3: CBC parameters, including hemoglobin, WBC, neutrophils, lymphocytes, eosinophils, and platelets, and their associations with disease severity and outcomes. t-test analysis with mean ± SD for CBC between each of survived and deceased, non-severe and severe COVID-19 cases. p ≤0.05 is significant. CBC - HB: Normal male 13.2-16.6 g/dL, Normal female 11.6-15 g/dL; WBC: 4.5-11.0 × 10⁹/L; RBC: Normal male 4.35-5.65 × 10¹²/L, Normal female 3.92-5.13 × 10¹²/L; HCT: Normal male 40-54%, Normal female 36-48%; Neutrophils: 2.0-8.0 × 10⁹/L; Eosinophils: 0.02-0.5 × 10⁹/L; Lymphocytes: 1.0-4.0 × 10⁹/L; Platelet: 150-400 × 10⁹/L. WBC: White blood corpuscles, RBC: Red blood corpuscles, HB: Haemoglobin, CBC: Complete blood count, SD: Standard deviation, Ns: Non-significant, HCT: Hematocrit.

Export to PPT

Conversely, WBC count was higher in severe cases (10.31 ± 2.59 ×10⁹/L) than in non-severe cases (9.78 ± 1.95 ×10⁹/L, p = 0.001), and even more elevated in deceased cases (15.3 ± 3.5 ×10⁹/L) compared to survivors (11.8 ± 2.1 ×10⁹/L, p <0.0001). NCs showed a similar increase in severe (7.08 ± 2.18 ×10⁹/L) versus non-severe cases (5.31 ± 1.23 ×10⁹/L, p <0.001) and in deceased (6.85 ± 1.20 ×10⁹/L) compared to survivors (5.50 ± 0.90 ×10⁹/L, p <0.0001). Lymphocyte levels were significantly lower in severe cases (0.75 ± 0.38 ×10⁹/L) than in non-severe cases (1.31 ± 0.46 ×10⁹/L, p <0.001), and even more so in deceased cases (0.71 ± 0.32 ×10⁹/L) versus survivors (1.50 ± 0.43 ×10⁹/L, p <0.0001). Eosinophils were slightly reduced in severe (0.02 ± 0.01 ×10⁹/L) compared to non-severe cases (0.04 ± 0.02 ×10⁹/L, p = 0.07) and significantly lower in deceased (0.02 ± 0.01 ×10⁹/L) compared to survivors (0.05 ± 0.01 ×10⁹/L, p <0.0001). Platelet counts were notably lower in severe cases (184.8 ± 58.8 ×10⁹/L) versus non-severe cases (256.4 ± 68.6 ×10⁹/L, p <0.001) and in deceased (213 ± 78 ×10⁹/L) compared to survivors (264 ± 92 ×10⁹/L, p <0.0001) [Figure 3].

DISCUSSION

This study evaluated key determinants of COVID-19 outcomes, focusing on demographic factors, comorbidities, and laboratory biomarkers. The findings confirm that advanced age, sex, and pre-existing conditions such as hypertension, diabetes, and asthma significantly increase disease severity and mortality risk.^[23] The overall mortality rate observed aligns with global estimates of ∼3.5%.^[12,24] Vaccination markedly reduced mortality, as reported internationally, with fatality rates ranging from 3.2% to 12.7% depending on vaccination status; in hospitalised cohorts, deaths occurred in 5.1% of vaccinated versus 8.3% of unvaccinated patients.^[25-27]

Our study examined the role of comorbidities such as hypertension, diabetes, asthma, and others in influencing COVID-19 mortality rates. We observed that hypertension (41%), diabetes (42.3%), and asthma (85.7%) were significantly associated with higher mortality rates among patients with COVID-19. These comorbidities strongly impacted disease outcomes, with low p values (<0.0001), indicating a statistically significant correlation. Our analysis aligns with previous studies that consistently report a higher risk of severe outcomes in patients with COVID-19 with comorbidities. For instance, Khedr et al.^[28] examined the impact of comorbidities on COVID-19 outcomes, emphasising the significant role of conditions such as cardiovascular disease (CVD) and diabetes in increasing disease severity. Their study, which analysed 439 patients admitted in mid-2020, found that 61.7% had comorbidities. Patients with pre-existing conditions, particularly CVD and diabetes, experienced worse symptoms, abnormal laboratory results, and higher rates of ICU admission, mechanical ventilation, and mortality. Recovery rates were substantially lower, and mortality rates were markedly higher among patients with comorbidities than those without.

In a study by Sanyaolu et al.,^[29] the significant impact of comorbidities on patient outcomes and the worsening of disease severity was highlighted. In the United States, hypertension was the most prevalent comorbidity, present in 55.4% of cases, followed by diabetes at 37.3%. Other notable conditions included hyperlipidemia (18.5%), coronary artery disease (12.4%), and renal disease (11.0%). Additionally, atrial fibrillation and heart failure (HF) were each reported in 7.1% of deaths. These findings underscore the heightened vulnerability of individuals with these conditions to severe COVID-19 outcomes, emphasising the critical need for targeted prevention and management strategies. Similarly, another study demonstrated that COVID-19 significantly increases risks for individuals with comorbidities, amplifying disease severity and mortality. Patients with diabetes, accounting for 11-58% of COVID-19 cases, face a higher risk of severe complications, with a fatality rate of 8% and ICU admission rates 14.2% higher than those of patients without diabetes. Hypertension, reported in 23% of cases in China, is linked to more severe infections and an increased risk of respiratory failure. CVD was present in 6.8-17% of non-survivors, contributing to complications such as ischemia and thrombosis. Additionally, kidney-related issues, observed in 3-9% of cases, included acute kidney injury (AKI) and proteinuria.^[30]

COVID-19 can cause both acute and long-term neurological complications.^[31] Research indicates that endothelial dysfunction and immune-mediated inflammation, more so than direct viral invasion, are the primary causes of this neural injury.^[32] Recognising these mechanisms is crucial for early patient identification and refining neuro-prognostic models.^[33]

Elevated D-dimer levels, a marker of hypercoagulability, have been linked to an increased risk of ischemic strokes and microvascular thrombosis in COVID-19, potentially leading to neurological deficits. Similarly, high CRP levels, indicative of a severe systemic inflammatory response, may exacerbate neuroinflammation and neurodegeneration, contributing to cognitive impairment and long-term neurological complications. Ferritin, a key marker of immune activation and oxidative stress, has also been associated with neuronal damage in critically ill patients with COVID-19, suggesting a possible link between inflammation-driven iron dysregulation and neurotoxicity.^[34,35]

In our study, D-dimer levels were found to be considerably greater in deceased patients compared to survivors, and in severe cases compared to non-severe cases. Our results align with previous studies,^[36-38] which have shown that patients with severe COVID-19 had distinctly higher D-dimer levels compared to those with mild cases, with median levels of 1.8 μg/mL in severe cases and 0.5 μg/mL in mild cases. A meta-analysis of 13 studies confirmed that D-dimer levels above 0.5 μg/mL are strongly associated with a higher risk of severe disease, with patients exhibiting such levels being nearly six times more likely to develop severe COVID-19, suggesting that elevated D-dimer levels may help identify patients at greater risk of severe illness.^[39] Our results were supported by another study examining the link between D-dimer levels and mortality in hospitalised patients with COVID-19. Six studies involving 1,355 patients were included, and it was found that D-dimer levels were remarkably higher in patients who died (average 3.59 μg/L) compared to survivors, with a strong statistical association (p <0.00001).^[40] In another study, D-dimer levels on admission were analysed, and it was found that a level greater than 2.0 µg/mL was a strong predictor of death during hospitalisation, with patients having a 51.5 times higher risk of dying compared to those with lower levels.^[41-45] These findings suggest that early measurement of D-dimer levels in patients with COVID-19 can help spot those at greater risk of severe outcomes and improve patient management.

In our study, CRP and ferritin levels were significantly higher in deceased and severe patients with COVID-19 than in survivors and non-severe cases, with p <0.0001 and 0.001, indicating their strong association with disease severity and mortality. These findings are consistent with previous research investigating the relationship between biomarkers such as CRP, D-dimer, and ferritin and COVID-19 severity.^[44,45] For example, elevated CRP levels were found to be significantly associated with worse outcomes, with a risk ratio (RR) of 1.84 (p <0.001). D-dimer levels were similarly linked to an increased risk, with a RR of 2.93 (p <0.001). Additionally, higher serum ferritin levels were observed in patients with poor outcomes, showing a standardised mean difference of 0.90 (p <0.0001).^[46]

In a study, significant increases in ferritin, D-dimer, and CRP were observed in critically ill patients. The severity groups’ differences were highly significant (p <0.001). It was noted that patients aged 50-60 were more likely to experience severe illness compared to younger patients, while sex did not significantly affect the severity (p >0.05). The biochemical markers (D-dimer, ferritin, and CRP) were strongly linked to the severity of COVID-19 symptoms.^[47] In another study, it was observed that D-dimer levels increased with age (p <0.0001). Age, cholesterol, fats, kidney function, inflammation, hemoglobin, and weight index were linked to higher D-dimer levels.^[48]

In this study, deceased patients had significantly higher PT levels (p <0.0001) than survivors, and severe cases had higher PT values than non-severe cases (p <0.0001). Patients with elevated PT, PTT, and INR were 2.4, 2.1, and 2.3 times more likely to require ICU admission, respectively. Various previous studies have highlighted the importance of monitoring inflammatory and coagulation biomarkers for early intervention in severe COVID-19 cases.^[49,50] It was observed that nearly half of the patients had prolonged PT, and >50% of severe and critically ill patients had both prolonged PT and elevated INR, indicating a higher risk of clotting problems. Thrombocytopenia (low platelet count) was found in 22.1% of patients, and it was more commonly observed in patients aged ≥55 years, with over half of them experiencing either low platelet counts or prolonged APTT.^[51]

In our study, we found that severe COVID-19 cases exhibited significantly lower hemoglobin and RBC counts compared to non-severe cases, while WBC, neutrophil, and platelet counts were elevated in severe cases. These findings align with previous studies, which reported higher WBC counts (over 7.70 × 10⁹/L), elevated NCs (over 5.93 × 10⁹/L), increased CRP levels (over 75.07 mg/L), and lower LCs in severe COVID-19 cases, along with higher levels of D-dimer, fibrinogen, and NLR, all of which were associated with poorer outcomes.^[52] Early identification and appropriate isolation of laboratory markers, including complete blood count profiles, in COVID-19 patients are essential for timely treatment, resource conservation, and the protection of both patients and medical staff, ultimately helping to prevent virus transmission in healthcare settings. In our study, LCs were found to be significantly lower in severe cases compared to non-severe cases, and eosinophil levels were also reduced, both of which correlated with worse outcomes. Additionally, critical patients exhibited lower levels of lymphocytes and eosinophils, while markers such as PT, D-dimer, and fibrin degradation products were elevated in more severe cases, as observed in other research analyses.^[53-55] These findings suggest that blood tests, including platelet count, PT, D-dimer, and NLR, can serve as valuable prognostic tools for predicting the severity and risk of poor outcomes in patients with COVID-19.^[55]

This study has several limitations that should be considered when interpreting the findings. The retrospective cohort design, while valuable for identifying associations between biomarkers and COVID-19 outcomes, restricts the ability to infer causality and may be influenced by biases inherent in data collection from medical records. Although statistical comparisons using chi-square and t-tests were appropriate for the large sample size, formal assessments of normality, variance homogeneity, and potential confounding were not fully addressed. Multivariate regression analyses were not performed due to missing data, sample imbalance between survivors and deceased cases, and concerns about model overfitting; consequently, residual confounding cannot be excluded. Despite these constraints, the consistency of the associations observed with previous studies supports the reliability of the results. Future studies should incorporate multivariate modelling, propensity score matching, and longitudinal designs to strengthen causal interpretation and control for confounding variables. Moreover, while parametric tests are generally robust in large samples, future work could include non-parametric alternatives such as the Mann-Whitney U test and Welch’s t-test to confirm statistical robustness. Finally, although the biomarkers examined, D-dimer, CRP, ferritin, and coagulation parameters, showed clear clinical relevance, establishing validated cutoff points for clinical application will require multicentre prospective studies with standardised protocols. Addressing these limitations will help refine biomarker-based risk stratification and enhance the predictive accuracy of COVID-19 outcome models.

CONCLUSION

This study has provided valuable insights into the complex interplay between demographic factors, comorbidities, laboratory markers, and COVID-19 outcomes. The high recovery rate underscores the effectiveness of current treatment strategies; however, the identified disparities and risk factors highlight the need for personalised care approaches. The significant role of comorbidities, such as hypertension, diabetes, and asthma, in influencing mortality rates calls for integrated management strategies for patients with pre-existing health conditions. Laboratory biomarkers, including inflammatory markers (D-dimer, CRP, and Ferritin) and coagulation profiles (PT, PTT, and INR), have emerged as critical tools for assessing disease severity and guiding clinical decision-making. The results of this study may serve as valuable biomarkers in the early management of high-risk patients with COVID-19, potentially improving prognosis and reducing mortality rates. Furthermore, these laboratory biomarkers may contribute to the development of policies and response strategies aimed at mitigating critical adverse outcomes of COVID-19.