Ferritin level: A predictor of severity and mortality in hospitalized COVID‐19 patients

Аbstrасt Introduction This study aims to investigate in‐hоsрitаl mоrtаlity in severe асute resрirаtоry syndrоme соrоnаvirus 2 раtients strаtified by serum ferritin levels. Methods Patients were stratified based on ferritin levels (ferritin levels ≤ 1000 or >1000). Results Approximately 89% (118) of the patients with ferritin levels > 1000 had pneumonia, and 51% (67) had hypertension. Fever (97, 73.5%) and shortness of breath (80, 61%) were two major symptoms among the patients in this group. Logistic regression analysis indicated that ferritin level (odds ratio [OR] = 0.36, 95% confidence interval [CI] = 0.21–0.62; p < .001), male sex (OR = 2.63, 95% CI = 1.43–5.06; p = .003), hypertension (OR = 4.16, 95% CI = 2.42–7.36; p < .001) and pneumonia (OR = 8.48, 95% CI = 3.02–35.45; p < .001) had significance in predicting in‐hospital mortality. Additionally, the Cox proportional hazards analysis and Kaplan–Meier survival probability plot showed a higher mortality rate among patients with ferritin levels > 1000. Conclusion In this study, higher levels of serum ferritin were found to be an independent predictor of in‐hоsрitаl mоrtаlity.

Conclusion: In this study, higher levels of serum ferritin were found to be an independent predictor of in-hоsрitаl mоrtаlity.

| INTRО DUСTIОN
In severe асute resрirаtоry syndrоme соrоnаvirus 2 (SARS-CoV-2) risk assessment, ferritin can be used as a biomarker to assess severity and mortality. 1,2 In SARS-CoV-2 infection, cytokine storms are interlinked with elevated levels of ferritin. High ferritin levels can cause pro-inflammatory changes and immunosuppression. 3 It was found that most diabetic SARS-CoV-2 patients who were critically ill had higher levels of ferritin. 4 Many studies have shown doubling of the ferritin level in elderly individuals, especially when they are older than 65 years compared to those aged younger than 50 years. 5 A multicentre study of SARS-CoV-2 infection reported a higher incidence of acute respiratory distress syndrome (ARDS), and increased morbidity was associated with higher hyperferritinemia. 6

| Stаtistiсаl аnаlysis
Descriptive statistics were used to summarize the data in the form of frequency, percentage, mean ± standard deviation (SD), and median ± interquartile range (IQR). Differences in patients with respect to study variables in the ferritin group were examined using the Pearson χ2 test. Logistic regression analysis was employed to check the effects of some study variables on cumulative allcause mortality. The Cox proportional hazards regression model and Kaplan-Meier survival were used to check how ferritin affected the mortality level. A 5% significance level was used to test the results. Statistical analyses were performed using SPSS version 27 (SPSS) and R software. 9
Logistic regression analysis showed that ferritin level, sex, hypertension, and pneumonia were significant predictors of all-cause cumulative mortality. A Cox proportional hazards analysis was conducted to determine whether ferritin had a significant effect on the hazard of mortality ( Table 6). The findings (LL = 9.85, df = 1, p = .002) show that ferritin was able to adequately predict the hazard of mortality. It is evident that at any particular time, patients with ferritin levels ≤ 1000 had a hazard that was 0.49 times as large as that of patients with ferritin levels > 1000 (B = −0.72, SE = 0.23, HR = 0.49, p = .001).

| 1651
A Kaplan-Meier survival probability plot was also included for ferritin. The plot represents the survival probabilities for different groups over time and shows that in the initial and later periods, the cumulative probability of dying was higher among patients with ferritin levels > 1000, but in the middle period, little difference was observed in the mortality rate of patients in the different ferritin groups (Figure 1).

| DISСUSSIОN
The main finding of our study is that the higher mortality rate among patients having ferritin levels > 1000. Other than ferritin levels gender, hypertension and pneumonia were found to be a predictor of in-hospital mortality. Around 89% of the patients having ferritin levels > 1000 had pneumonia and 51% had hypertension. The mean age of the patients was 54.0 ± 15.3 years and among which the ratio of male to female was 233:362. A male predominance was noted in the group with ferritin levels > 1000. Higher levels of C-Reactive Protein and Procalcitonin were seen in ferritin > 1000.
Serum ferritin was found to be an independent predictor of severe SARS-CoV-2 disease. 10 A study from Indonesia Rasyid et al. documented a mean ferritin level of 1689 in critically ill SARS-CoV-2infected patients. 11 SARS-CoV-2 patients with cytokine storm were also found to have significantly higher levels of ferritin. 12 Several autopsies of SARS-CoV-2 patients revealed higher ferritin levels. 13 Elderly SARS-CoV-2 patients with elevated ferritin levels showed higher mortality than those with lower ferritin values. 11 In another study, the incidence of ARDS was higher in those with hyperferritinemia. 14 T A B L E 3 Laboratory findings of COVID-19 patients grouped by ferritin level (ng/ml)    15 Elevated ferritin levels can be used as a biomarker to stratify high-risk patients from low-risk patients, which may in turn help in the early identification and management of SARS-CoV-2 patients. 16 Hyperferritinemia was more common in critically ill and discharged SARS-CoV-2 patients than in stable hospitalized patients. 17 Unlike our study, hypertensive SARS-CoV-2 patients had lower levels of serum ferritin, as reported by Huang et al. 18 Similar to our study, a study by Phipps et al. 19 showed that the severity of acute liver failure in SARS-CoV-2 patients was more common in patients with hyperferritinemia. The frequency of ICU admission was higher in SARS-CoV-2 patients with hyperferritinemia. 20 Similar findings were also reported in our study. In another study, SARS-CoV-2 patients with cancer had higher serum ferritin levels than those without cancer. 21 The clinical association of hyperferritinemia in SARS-CoV-2 in terms of mortality, comorbidities, and severity was well established in a meta-analysis. 22

| LIMITATIONS
Our study has various limitations. Its retrospective design limits causal inference. Unmeasured confounding factors, such as clinical comorbidities and medications, could have affected the outcomes. This Kuwaiti study included all SARS-CoV-2-positive patients and undoubtedly consisted of mainly milder cases of the disease.
However, if it included SARS-CoV-2 patients who typically consist of a significant case mix of mechanically ventilated and critical cases of patients, the findings might have looked different.

| CО NСLUSIОNS
This study demonstrated that hyperferritinemia is an independent predictor of in-hоsрitаl mоrtаlity in SARS-CoV-2 patients. The incidence of ICU admission was higher with hyperferritinemia. More prospective studies are required to better understand hyperferritinemia and in-hospital mortality in SARS-CoV-2.

CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.

DATA AVAILABILITY STATEMENT
The data that support the results of the study are available on request from the corresponding author.
F I G U R E 1 Kaplan-Meier survival plot of mortality grouped by fertility