COVID-19 with high-sensitivity CRP associated with worse dynamic clinical parameters and outcomes

Objective This study aimed to evaluate the relationship between high-sensitivity C-reactive protein (hsCRP) in hospitalized COVID-19 patients and their clinical outcomes, including trajectory of hsCRP changes during hospitalization. Method and results Patients with positive COVID-19 tests between 2021 and 2023 were admitted to two hospitals. Among 184 adult patients, approximately half (47.3%) had elevated hsCRP levels upon admission, which defined as exceeding the laboratory-specific upper limit of test (> 5.0 mg/L). Clinical outcomes included critical illness, acute kidney injury, thrombotic events, intensive care unit (ICU) requirement, and death during hospitalization. Elevated hsCRP levels had a higher risk of ICU requirement than those with normal, 39.1% versus 16.5%; adjusted odds ratio (aOR), 2.3 [95% CI, 1.05–5.01]; p = 0.036. Patients with extremely high (≥2 times) hsCRP levels had aOR, 2.65 [95% CI, 1.09–6.45]; p < 0.001. On the fifth day hospitalization, patients with high hsCRP levels associated with acute kidney injury (aOR, 4.13 [95% CI, 1.30–13.08]; p = 0.016), ICU requirement (aOR, 2.67 [95%CI, 1.02–6.99]; p = 0.044), or death (aOR, 4.24 [95% CI, 1.38-12.99]; p = 0.011). The likelihood of worse clinical outcomes increased as hsCRP levels rose; patients with elevated hsCRP had lower overall survival rate than those with normal (p = 0.02). The subset of high hsCRP patients with high viral load also had a shorter half-life compared to those with normal hsCRP level (p = 0.003). Conclusion Elevated hsCRP levels were found to be a significant predictor of ICU requirement, acute kidney injury, or death within 5 days after hospitalization in COVID-19 patients. This emphasized the importance of providing more intensive care management to patients with elevated hsCRP.


Introduction
Coronavirus disease (COVID- 19), identified first in December 2019, is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), an enveloped, positive-sense, single-stranded RNA virus belonging to the Betacoronavirus genus of the Orthocoronavirinae subfamily in the Coronaviridae family.Six types of coronaviruses (CoV-229, CoV-OC43, CoV-NL63, CoV-HKU1, SARS-CoV, and MERS-CoV) can cause human respiratory tract infection (1).The primary mode of transmission is person-to-person via direct contact and respiratory droplets.COVID-19 symptoms range from mild or undetectable symptoms to highly severe symptoms, including death (2).Disease severity and mortality rates are closely associated with comorbidities such as advanced age, diabetes, hypertension, and cardiovascular diseases (3).Several inflammation markers, including cytokines, nitrogen species, and mediators, have been documented to predict severity (4), and these markers widely use C-reactive protein (CRP) and high-sensitivity CRP (hsCRP) (5)(6)(7).Elevated levels of acute-phase proteins are associated with various cytokines, which indicate inflammation.Worse COVID-19 clinical outcomes are caused by cytokine storms and the formation of widespread microthrombi among multiple organ systems (8).Novel biomarkers are being investigated to predict clinical outcomes (9).The hsCRP test is widely used to assess cardiovascular disease risk or monitor inflammation in patients because hsCRP is more sensitive than CRP detection, which detects vessel inflammation.Theoretically, the COVID-19 virus can damage respiratory and vascular tissues.Several studies have demonstrated that elevated hsCRP levels are correlated with worse clinical outcomes in patients with COVID-19 infection, suggesting that hsCRP might be a valuable biomarker for predicting the severity of clinical outcomes (8,10).However, the correlation between hsCRP and hospitalization is not fully understood.The dynamics of hsCRP levels in patients with COVID-19 infection are still being investigated.We hypothesized that hsCRP is associated with worse clinical outcomes among hospitalized COVID-19 patients.This study analyzed clinical data from two hospitals in Thailand to explore the association of hsCRP levels with all-cause mortality and the likelihood of hospital discharge.

Patient cohort and demographic data
This retrospective observational study was conducted at the Rajavithi and Rangsit (Rajavithi-2) Hospital.The study enrolled 184 adult patients in Thailand between 2021 and 2023.All patients had tested positive for SARS-CoV-2, confirmed by real-time polymerase chain reaction (RT-PCR) upon admission to the hospital.Pregnant women and patients who were chronically using anti-inflammatory drugs were excluded from the study.Demographic information was collected from the medical records of all adult patients (over 18 years old).Laboratory tests, including a complete blood count, blood chemistry panel (including renal and liver function tests), coagulation parameters, and measurements of acute phase reactants (including ferritin, ferritin, and D-dimer), were conducted.Radiological chest X-rays and computed tomography (CT) scans were performed.The Ethics Committee of Rajavithi Hospital approved this study (Number 198/2564).

Molecular testing for SAR-CoV-2 by RT-PCR; Ct value method
Patients admitted with positive COVID-19 RT-PCR results were enrolled retrospectively.The Bio-Rad CFX96™ RT-PCR instrument constructs a real-time amplification curve based on the signal changes.The qualitative detection of the SARS-CoV-2 novel coronavirus at the nucleic acid level was reported in the FAM channel for ORF1ab of Ct value, N (labeled by VIC).A Ct value less than 25 was considered a low SARS-CoV-2 viral load (11,12).High-sensitivity C-reactive protein (hsCRP) levels were measured using the Abbott Architect c16000 automated biochemical analyzer (Abbott Diagnostics, North Chicago, USA).The reference range for hsCRP was 0-5.0 mg/L.Adults with COVID-19 were categorized into two groups based on hsCRP levels at admission: those with values exceeding the referenced high cutoff of 5 mg/L were designated as the "high hsCRP group, " while all others were classified as the "low hsCRP group." Measurements were tested on days 0, 3, 5, 7, 10, and 14.

Clinical severity classifications
COVID-19 severity was classified as follows: moderate cases were considered to have clinical or radiological signs of lower respiratory disease and an oxygen saturation (SpO 2 ) level of at least 94% on room air.Severe cases were defined as a SpO 2 level of less than 94%, a respiratory rate of more than 30 breaths/min, or radiological findings of lung infiltrates of more than 50% of the chest X-ray.Critical cases included patients who presented with respiratory failure, septic shock, or multiple organ dysfunction.Complications included acute kidney injury and thrombosis events.Clinical outcomes were admission to the intensive care unit (ICU) or death during hospitalization.

Statistical analysis
Categorical variables are presented as numbers and percentages with comparison by χ 2 tests.Continuous variables were characterized by medians with interquartile ranges (IQRs) for non-normally distributed data.A Mann-Whitney U-test was performed for all non-normally distributed data.The longitudinal trajectory of the mean hsCRP level per day during hospitalization for all patients in each clinical outcome category is shown using the fitted values from the general linear model for each time point separately.Logistic regression models to estimate the odds of different clinical outcomes with the covariates in the multivariable models include age, sex, type 2 diabetes, dyspnea, and initial laboratory results for neutrophillymphocyte ratio (NLR), alanine transaminase (ALT), and estimated glomerular infiltration rate (eGFR).Kaplan-Meier curves were performed to estimate the overall survival rate.Statistical analysis was performed using SPSS program version 22.0 (Mahidol University license), with a significance level set at a p-value of ≤0.05.

HsCRP and all-cause mortality
Among the 184 hospitalized patients with COVID-19, 65 (35.3%)died, and 119 (64.7%) were discharged.Adjusted mortality was higher among patients with versus without elevated hsCRP levels on day 5 after hospitalization (aOR 4.24 [95% CI, 1.38-12.99];p = 0.011).The Baseline hsCRP levels and severe conditions during hospitalization.aOR, adjusted odds ratio, *p < 0.05.2).Among 87 (47.3%) patients with elevated hsCRP levels at admission, the Kaplan-Meier survival curve demonstrated a significant difference between patients with elevated hsCRP and those    without (log-rank test; p = 0.02).Similar to the subgroup analysis of patients with high viral load (Ct value <25) and elevated hsCRP, 93 (50.5%) high viral load patients had elevated hsCRP.The Kaplan-Meier curve showed a significantly lower overall survival rate in patients with elevated hsCRP than in those without (p = 0.003) (Figure 4).

Discussion
Adult hospitalized with COVID-19 infection at two hospitals in Bangkok, Thailand, 47.3% had an elevated hsCRP level.Those patients with initially elevated hsCRP showed a 2-fold increased risk of ICU admission requirement.Interestingly, of serial hsCRP during  hospitalization, those who tested for hsCRP on day 5 were associated with a 4-fold increased risk of acute kidney injury events.To the best of our knowledge, this study is the first to investigate the impact of dynamic hsCRP among patients with COVID-19 infection.Serial follow-up hsCRP levels within 5 days after hospitalization also showed an independent association with the increased requirement of an intensive hospital setting as well as death.The requirement for intensive care and mortality exhibited a dose-dependent correlation with the third hsCRP measurement in these patients.This finding is similar to the study of CRP concentration to show the association between respiratory failure and cardiovascular dysfunctions (13,14).The patients were followed and categorized two times high hsCRP levels; those patients also showed between 4-and 7-fold increased ICU setting requirements.Patients with high hsCRP had a 3-to 4-fold increased risk of mortality (7).Our findings might be used to triage patients for immediate treatment and tailor treatment plans to individual patient needs.Interestingly, patients with high hsCRP levels had shorter survival rates, especially those with a high viral load.The present study suggests the association between patients with either high hsCRP or SAR-CoV-2 viral loads and high mortality due to high viral burden and high inflammation conditions (15)(16)(17)(18) This study adds to this evidence by demonstrating that hsCRP levels at admission possess similar discriminatory ability, suggesting its potential utility as a readily available and cost-effective marker.Most studies that used CRP demonstrated that this test is related to increased mortality rates due to the viral destruction of the vascular injury process.HsCRP is one of the most popular tests related to vascular complications.Our patient also had high acute kidney injury within 5 days after admission, which was explained by vascular injury events (14).However, small thrombotic events have been documented, especially in patients with preexisting diabetes and liver disease (24,25).This is a limitation of the COVID-19 patient's autopsy investigation because it is a highly contagious disease.
The limitation of this study was that it was a retrospective study, and the hsCRP test was not routinely and continuously performed in all patients.Some patients were excluded during data collection.Second, cardiovascular events depend on the diagnosis conducted by the attending physicians.However, our data were validated by two independent physicians.The bias or discordance is unknown.Third, no autopsy was performed because of the easy spread of viral properties.This may miss some data on death causes.Finally, we did not assess subsequent cardiovascular injury events among discharge patients.

Conclusion
This study adds to the growing evidence that elevated hsCRP levels measured at admission significantly differentiate COVID-19 patients with poor clinical outcomes, such as needing ICU care or dying.Additionally, rising hsCRP levels within 5 days of hospitalization independently predicted these poor outcomes, even when accounting for vascular injury events like acute kidney injury.Notably, patients with high hsCRP and exceptionally high viral loads had shorter life expectancies.These findings support the potential of using hsCRP as an additional biomarker to represent a direct correlation with the severity of COVID-19 infection and predict patient outcomes.

FIGURE 4
FIGURE 4 Kaplan-Meier curves to assess survival among patients with COVID-19 during 28-day hospitalization.The curves are color-coded stratifying to hsCRP levels (elevated hsCRP-blue blonde line, normal hsCRP-black dotted line).(A) Represents all patients and (B) represents only high COVID-19 viral test (the Ct value of <25).

TABLE 1
Baseline characteristics and clinical data of patients with COVID-19 stratified by HsCRP at admission.

TABLE 2
Different HsCRP categories with unadjusted and adjusted OR for predicting patient outcome.
(19)erefore, the hsCRP level might guide the physician in estimating a prognosis and identifying poor prognosis patients with COVID-19 infection, especially new variants.The top four clinical presentations were fever, cough, dyspnea, and sore throat.The study by Lacobucci G et al. also showed that the omicron variant of COVID-19 had four common symptoms at admission(19).This epidemic data correlated with demographic data on the spread of COVID-19 in Thailand during 2021-2022 (20).