Serum ferritin level during hospitalization is associated with Brain … – Nature.com

Participants and surveys of residual symptoms

Higashiosaka City Medical Center is the central public hospital of Higashi-Osaka City, a city in the Kinki region of Japan, with a total population of approximately 500,000. This hospital is a tertiary medical center in Osaka Prefecture offering treatment for COVID-19. All patients were diagnosed with COVID-19 by positive nasopharyngeal swab polymerase chain reaction (PCR). In general, patients admitted to our hospital were either elderly or at a high risk of severe respiratory failure; however, patients who required tracheal intubation were transferred to other hospitals that provided intensive care for COVID-19 respiratory failure. After receiving acute medication for COVID-19, they were discharged or transferred to rehabilitation hospitals, long-term care hospitals, or intensive care hospitals for tracheal intubation management for severe respiratory failure.

We surveyed post-COVID syndrome by sending questionnaires. Questionnaires were mailed to patients at least 5 months after admission. Responses to the question, Do you feel foggy or unfocused 1 month after onset of COVID-19? and Do you feel foggy or unfocused 3 months after onset of COVID-19? were categorized as 0=never, 1=rarely, 2=occasionally, 3=frequently, or 4=always. This study included patients admitted to our hospital between October 10, 2020, and October 31, 2021. The participants were categorized into three groups according to their admission date: third wave, October 10, 2020, to February 28, 2021 (surveyed on May 10, 2022); fourth wave, March 1, 2021, to June 20, 2021 (surveyed on May 10, 2022); and fifth wave, June 21, 2021, to October 31, 2021 (surveyed on March 25, 2022). Questionnaires were not sent to patients who died at our hospital. Questionnaires were sent to 1061 patients, 404 of whom responded. We included 253 patients 1 month after onset of COVID-19 in the analysis, excluding those hospitalized for 1 day, who had no data on ferritin, C-reactive protein (CRP), or white blood cell counts, and those who left the Brain Fog question blank (Fig.1). We included 229 patients when the data 3 months after the onset was used, which was fewer than 1 month as more patients left the Brain Fog question blank.

The following data were collected from the medical records, namely, age, sex, days in the hospital, length of hospitalization, tracheal intubation or not, peak white blood cell count, red blood cell count, platelet count, and levels of albumin, creatinine, Na, K, peak CRP, D-dimer, procalcitonin, HbA1c and treatment.

The questionnaire was used to obtain Brain Fog scores 1 month after COVID-19 infection onset (No symptoms:0, few:1, sometimes:2, frequently:3, always:4). Patient characteristics and ferritin levels were compared in the presence and absence of any Brain Fog. Unless otherwise specified, categorical variables were tested using a Chi-Squared test and continuous MannWhitney U or Students t test. Furthermore, the peak ferritin level of each Brain Fog-point group was analyzed using Linear Regression.

Propensity score analyses were conducted to exclude confounders as far as possible. Some variables (Model 1; age, sex, Model 2; age, sex, CRP, and WBC, Model 3; age, sex, CRP, WBC, length of hospitalization, intubation, and maximal oxygen dose, Model 4; age, sex, CRP, WBC, length of hospitalization, intubation, maximal oxygen dose, D-dimer, creatinine, sodium, potassium, albumin, red blood cell count, platelet counts, and procalcitonin values, Model5; age, sex, CRP, WBC, length of hospitalization, intubation, maximal oxygen dose, D-dimer, creatinine, sodium, potassium, albumin, red blood cell count, platelet counts, and procalcitonin values, remdesivir, steroid therapy, antibody cocktail therapy) were regressed by Logistic Regression using the presence of Brain Fog as a predictor. Each patient propensity score was calculated as the probability of being classified into one specified group. The variables of each model were selected for the following reason, Model 1; basic social background, Model 2; inflammation-related parameters, Model 3; factors used for the determination of severity, Model 4: factors including maker of bacterial inflammation, Model 5: factors including treatment. Following greedy pair matching, each group of patients was matched using the nearest propensity scores, and we compared the two groups using the Wilcoxon signed-rank test.

Finally, the contribution of ferritin levels to the difference among patients with and without Brain Fog was tested between each COVID-19 phase using a permutation test. Specifically, the ferritin values of individuals with Brain Fog and those without Brain Fog were randomly permuted within each phase, and the median difference values of those with Brain Fog and those without brain fog were calculated based on 5000 samples. Then, a one-sided p value was calculated as the proportion of sampled values where the true median difference was located. All analyses were performed using Python 3 (Python Software Foundation, Wilmington, DE, USA).

This study was conducted according to the guidelines of the Declaration of Helsinki on Research Involving Human Subjects. All methods were performed according to the relevant guidelines and regulations and observational studies. The Ethics Committee of Higashiosaka City Medical Center approved the study design and protocol and waived the need for the requirement of Informed Consent because of the retrospective nature of the study.

Excerpt from:

Serum ferritin level during hospitalization is associated with Brain ... - Nature.com