Discussion

In this case series, we report the clinical characteristics and risk factors associated with outcomes in COVID-19 patients admitted to Shariati hospital, Tehran, Iran. Our study comprehensively describes clinical presentations and laboratory findings alongside with radiologic features of a COVID-19 registry among the Iranian population. The study investigates whether patients who died of COVID-19 are different from those who recovered in the disease course.
Sex showed no correlation with the mortality rate in our study population. Deceased cases were significantly older compared to the surviving patients, which is in line with previous studies (T. Chen et al., 2020; Zhou et al., 2020). Although the difference in body mass index (BMI) was not statistically significant within the study subgroups (p=0.3436), Severe and morbidly obese patients were at higher risk of death. This was similar to a study by Simonnet et al. that showed a correlation between BMI above 35 and the severity of the disease, and also in-hospital death (Simonnet et al., 2020). Unfortunately, due to the unavailability of BMI in every patient, we were unable to analyze it further.
Within the presenting symptoms, impaired consciousness, ranging from mild disorientation to stupor, significantly increased mortality. Impaired consciousness showed a strong impact on the patients’ final outcome, affecting the moribund group 13 times more than the survived ones. In regards to neurologic manifestations, Mao et al. also observed 14.8% of their severe cases had impaired levels of consciousness (Mao et al., 2020).
Analyzing baseline vital signs values on admission led to impressive conclusions. Body temperature over 39 °C was strongly suggestive of mortality in our sample. We noticed systolic blood pressure (SBP) higher than 140 mmHg on admission lowers the mortality rate down to 60% while SBP below 100 mmHg increased the risk of mortality up to around three folds, which is inconsistent with prior observation indicating the association of higher levels SBP with severe clinical picture (Huang et al., 2020).
Besides, RR more than 25 and PR values over 100 were significantly associated with a higher mortality rate. Among them, PR remained statistically significant in our ultimate logistic regression model with multiple variants as well. This strong ability of a single component in physical examination to predict disease course is quite remarkable. There is a lack of correlation between PR patterns with other suggestive, inflammatory variables such as CRP and temperature in our multivariate analysis. Thus, we hypothesize that besides systemic inflammation responses, higher PR could arise from additional cardiac involvement by the virus independent of mere inflammation. Similarly, in a study by Zhou et. Al. PR ≥125 and RR >24 were associated with a higher risk of mortality, although none of the two variables remained in the multivariate analysis to assess their correlation to others (Zhou et al., 2020).
Analyzing the CT scan findings of the patients’ lung involvement revealed consolidative opacities as the significant imagining manifestation within our population. Bernheim et al. also suggest bilateral and peripheral ground-glass and consolidative pulmonary opacities as the hallmarks of COVID-19 infection on imaging (Bernheim et al., 2020).
History of HTN prior to hospital admission did not significantly increase the risk of mortality in our study population. This observation is inconsistent with prior studies reporting the adverse role of HTN in the COVID-19 course. HTN is repeatedly reported as the most prevalent comorbidity among COVID-19 patients (Huang et al., 2020; Wu et al., 2020). Moreover, a recent meta-analysis by Lippi et al. concluded that HTN is associated with a 2.5-fold higher risk of severe disease course or mortality in SARS-Cov-2 infections (Lippi, Wong, & Henry, 2020). Probably our small number of subjects has not allowed us to detect this association as evident by our wide 95%CI. In addition, it is not clear whether poor BP control had contributed to severe symptoms and morbidity among the hypertensive population, or the mortality was due to the higher frequency of HTN in the older population, which are more vulnerable to COVID-19 infection (Schiffrin, Flack, Ito, Muntner, & Webb, 2020).
We observed that venous blood acidosis was associated with increased mortality. Severe acidosis (pH of 7.25 or lower) led to 4 times higher mortality, and PCO2 below 35mmHg was associated with a 2.5-fold increase in mortality. Since no statistically significant correlation was detected between PCO2 and RR, their attribution to mortality is independent. In addition, blood bicarbonate contents below 22 mmol/L led to 2.6 times more mortality than higher levels.
Along with the existing claim on the higher incidence of in-hospital death by an increase in neutrophil to lymphocyte ratio (NLR), our data also revealed higher NLR in the deceased group compared to the recovered patients. Liu et al. concluded NLR as an independent risk factor of the in-hospital mortality, especially within their male candidates (Liu et al., 2020). Besides, Qin et al. noticed a statistically significant higher number of neutrophils alongside with a lower number of lymphocytes in severe to critical patients compared to non-severe cases. This describes the positive association of Neutrophilia or lymphopenia with disease severity and increased rate of death (Qin et al., 2020). Ultimately a recent meta-analysis by Lagunas-Rangel confirms the strong association between higher levels of NLR to death in COVID-19 patients (Lagunas-Rangel).
Since only 166 cases in our study population had their urine analyzed the consequent data is not strongly representative. However, serum creatinine levels alongside proteinuria were significantly detected higher in the deceased subgroup compared to the recovered patients. Based on a recent study by Xu et al., ACE2 expresses equally in kidney cells in comparison to that in the lung, esophagus, small intestine, and colon. This suggests that the kidney could be considered as an important target organ for SARS-CoV-2, but further studies should address this issue (Xu).
It is noteworthy that our study was conducted in a tertiary referral center which is among the most well-known and well-equipped centers in Tehran, especially in internal medicine subspecialties. Keeping this in mind, during the recent outbreak, the number of patients with long-term rheumatologic complaints was diagnosed with confirmed SARS-COV-2 infection far less than patient with other chronic diseases such as cancers in our center (11 and 40 respectively). Hence, we hypothesize that consuming anti-inflammatory drugs would protect the patients from the new coronavirus infection due to confirmed inflammatory responses caused by COVID-19.