Baseline Serum Total Cortisol During the Primary Corona Virus Infection in the Beginning of the Pandemic in Cameroon

Introduction: Improvement of critically ill patients Covid 19 positive placed on glucocorticoids may suggest impairment of the adrenal function. Our objective was to evaluate baseline serum cortisol in covid+ patients. Methods: This was a descriptive and analytical cross-sectional study conducted in a population of patients infected with 2019-nCoV in Cameroon. Patients receiving corticosteroids irrespective of the route of administration within 3 months prior to recruitment and those with any hypothalamic-pituitary-adrenal axis disease were excluded. We reviewed the medical records of patients to collect socio-demographic and clinical data. Patients were then sampled at 8 am for serum cortisol assay by the competitive ELISA method. Statistical analyses were performed using the Student's test to compare means. The signicance level was set at p<0.05. Results: We included 80 covid+ patients, predominantly males (45, 56.3%) with a mean age of 43±13 years. The co-morbidities found were hypertension (9, 11.3%), diabetes mellitus (7, 8.8%), cardiovascular disease (8, 10%) and obesity (4, 5%). The most common symptoms were asthenia (44, 55%), fever (36, 45%) and respiratory symptoms (75, 93.7%). The non-severe form was the most common (70, 87.5%). Only 12.5% (10) of patients received oxygen therapy within our study. The mean baseline serum cortisol in the series was 279.55 nmol/l±128.68 (min 96.1332 nmol/l; max 665.0521 nmol/l). Sixty-nine patients (86.3%) had baseline cortisol ≤ 413.79nmol/l, suggesting an insucient response to stress. No statistically signicant association was found between serum cortisol and disease severity. Conclusion: The absence of a marked rise of cortisol during COVID-19 suggests possible involvement of the hypothalamic-pituitary-adrenal axis in this infection.

peripheral resistance to glucocorticoids which is associated with a poor prognosis and requires supplementation (5). Few data exist on the adrenal response induced by COVID-19 infection. The purpose of this preliminary study was to assess baseline adrenal function during primary infection with the new coronavirus and to correlate it with disease severity.

Methods:
Setting and participants: The study was conducted in the isolation unit for the Covid positive patient at the Yaoundé central hospital. This is a unit created as part of the Covid pandemic since March 2020 when Cameroon recorded its rst positive case of Covid 19. The department has a capacity of 26 beds with an intensive care unit suitable for critical patients requiring oxygen.
Study design: We conducted a cross-sectional study of Covid-19 management centres in Cameroon.
Inclusion criteria: We included consenting participants followed up in covid-19 management centers with positive virological RT-PCR results.
Exclusion criteria: Patients on corticosteroids, regardless of the route of administration taken during the 3 months prior to recruitment, patients with hypothalamic-pituitary-adrenal axis pathology and pregnant women were excluded.
Procedure: Data were collected from the medical records of the patients concerned. Age, sex, comorbidities, clinical presentation, ambient oxygen saturation and resort oxygen therapy were recorded.
Blood samples were then taken at 8 am to respect the nycthemeral cycle of cortisol secretion after curbing the stress of direct venipuncture. The assays were performed by competitive ELISA in a serial manner according to laboratory procedures. The biological data were interpreted as follows: a cortisol threshold ≤ 413.79 nmol/l was considered to be an insu cient cortisolic response to infection and > 413.79 nmol/l normal.
Statistical analyses: Patients were then classi ed into 2 subgroups according to severity: subgroup with oxygen therapy and subgroup without oxygen therapy. Data were compiled and analyzed using SPSS version 23 software. Quantitative variables were preported as means and standard deviation. These were compared using student t-test. Qualitative variables were presented as percentages and these were compared with a chi-square test. The statistical signi cance level was set at 0.05.

Association between total baseline cortisol and clinical features of Covid-19
We did not nd a statistically signi cant association between serum cortisol and oxygen therapy, asthenia, comorbidities (hypertension, diabetes mellitus and obesity), number of patient symptoms and duration of symptoms for a P < 0.05 value. In contrast, serum cortisol level was associated with patient age. (Table 2) The objective of this study was to evaluate baseline total serum cortisol in covid positive patients. We therefore perfomed an 8 AM serum total cortisol in all patient positive to Covid 19 who were hospitalized in the isolation unit for Covid-19 patients. We found that the mean baseline total cortisol level was 279.5 nmol/l ± 128.7. The mean cortisol level was higher in the group with oxygen therapy versus no oxygen therapy (316.0 nmol/l ± 49.7 versus 124.54 nmol/l ± 14.9). Nine patients had patent adrenal insu ciency with serum cortisol ≤ 137.9 nmol/l. We did not nd a statistically signi cant association between serum cortisol and oxygen therapy. In contrast, serum cortisol level was associated with patient age.
The diagnosis of adrenal insu ciency requires ACTH stimulation test which is the most speci c test for the diagnosis of adrenal insu ciency. However, Covid-19 infection in itself could constitute a stimulating stress for the HPA axis by the induced in ammatory systemic response. As a preliminary study, baseline adrenal function should be assessed in all COVID + patients regardless of clinical disease status.
The technique used for the bioassay was the competitive ELISA method. Although this is not the reference method, it is easy to perform and suitable for screening. Its threshold of sensitivity is 88% and it is accessible in our context.
The socio-demographic and clinical characteristics of our study population are consistent with those found in the Wuhan cohort in China. Men are the most affected by the disease and have diabetes mellitus and hypertension as the main co-morbidities. The clinical presentation is by far dominated by asthenia and fever (6).
Males are believed to be most vulnerable to 2019-nCoV due to the greater expression of ACE receptors in men than in women. In addition, men are more susceptible to co-morbidities (hypertension, diabetes, cardiovascular disease) and therefore to severe forms of the disease (7).
Diabetes is an independent predictor of morbidity and mortality in COVID-19 (8,9). Regardless of the time of discovery, it predisposes to severe forms through its susceptibility to infection. This is due to impairment of immune mechanisms, dysfunction of immune cells, oxidative stress caused by hyperglycemia and hypersecretion of counter-regulatory hormones (10). However, SARS-CoV-2 as a risk factor for glycemic dysregulation is yet to be demonstrated. Some authors have raised the hypothesis of viral colonization of the pancreas through ACE receptors expressed in islets of Langerhans, however pancreatic viral particles have not been isolated in the autopsy series(8).
The mean baseline cortisol level (279.5 nmol/l ± 128.7) with 69 patients (86.3%) with baseline cortisol ≤ 413.79 nmol/l and 09 patients with cortisol less than 137.93 nmol/l represent an insu cient adrenal response to the stress induced by 2019-nCoV infection. Tricia Tan et al. in 2020 found an association between baseline cortisol threshold value ≥ 744 nmol/l and worsening morbidity/mortality in COVID-19. They compared the value of the patient's baseline cortisolaemia within 03 days following the onset of symptoms and its fate during the following 42 days. This suggests integrity and an adapted response of the corticotropic axis in the acute phase of infection (11).
Nevertheless, these results are not applicable in our context since our patients were enrolled later in the infection (beyond 03 days). The cytokine storm characteristic of SARS-CoV-2 infection and predictor of disease severity occurs from day 6 (2,12). This overly inadequate in ammatory response to viral aggression would also be incriminated in the syndrome of multivisceral failure with damage to the corticotropic axis in COVID-19 by destruction of the adrenal parenchyma(4). This suggests an alteration of the cortisolic response later in the infection. In our series, it is clear that the mean baseline cortisol level decreases with the duration of symptoms (293.3 nmol/l ± 115.2 vs 269.24 nmol/l ± 142.2).
The cortisolic response is inadequate in both the oxygen therapy group and the non-severe group. This suggests an insu ciency of basal cortisol secretion in response to stress in all phases of the disease, from less symptomatic patients to patients on respiratory assistance. During the SARS outbreak in 2003, adrenal insu ciency was described in patients infected with SARS-CoV 1 (4,13,14). Three mechanisms were incriminated: 1-Necrosis of the adrenal tissue found in autopsy series by viral colonization of the adrenals through ACE receptors expressed in this tissue and the cytokine tornado triggered by the viral infection and directed to the adrenals.
2-Viral in ltration of the hypothalamo-pituitary axis through ACE receptors expressed in this tissue and resulting in hypopituitarism.
3-Corticotropic insu ciency caused by antibodies to ACTH. This is explained by the molecular analogy between viral proteins and ACTH (15).
2019-nCoV, a virus of the same family as SARS-CoV-1, could also trigger the same mechanisms and be responsible for adrenal insu ciency.
Preliminary results from the RECOVERY study published in June 2020 showed a signi cant reduction in mortality in severe patients with dexamethasone (10). This potent corticosteroid is believed to be effective in reducing the cytokine storm characteristic of severe and lethal forms of COVID-19. Hence the bene t of corticosteroid therapy in severe forms of the disease which will act as an anti-in ammatory and could also have a bene t in hormone supplementation if there is associated cortisol de ciency. In our cohort, there is evidence of an insu cient and inadequate cortisol response that may require supplementation for effective control of COVID-19.

Conclusion:
The absence of a marked rise in serum cortisol during COVID-19 would suggest possible involvement of the hypothalamic-pituitary-adrenal axis during this infection.