Résumé
Hospitals were overburdened during peak periods of Coronavirus disease 2019 (COVID-19) pandemic, and bed occupancy was full. The ability to predict and plan patients' hospital length of stay allows predictability in terms of the free capacity of hospital facilities. The purpose of this article is to evaluate the factors that influence the hospital length of stay among discharged (recovered) from COVID-19 patients. This will allow the prediction of the likely number of bed days in the conditions of intensive workload of medical facilities for hospital care. A total of 441 discharged after hospital treatment for COVID-19 patients are followed up. Factors for prolonged hospital length of stay are searched among the indicators recorded at admission. Median hospital length of stay of the patients discharged from COVID-19 ward is 9 days (IQR 6-12) and in the COVID-19 intensive care unit 12 days (IQR 9.75-18.75). The median length of stay assessed by a survival analysis is 35 days in the COVID-19 unit and only 8 days in intensive care, due to the high mortality in the intensive care unit. The longer hospital length of stay of patients discharged from the COVID-19 wards is associated with the presence of hypertension (median 10 vs. 8 days for patients without the disease, p=0.006), ischemic heart disease (10 vs. 8 days, p<0.001), cerebrovascular disease (10 vs. 8 days, p=0.061 - did not reach significance), peripheral arterial disease (12 vs. 8 days, p=0.024), chronic renal failure or chroniodialysis (14 vs. 8 days, p<0.001), oncological illness (11 vs. 8 days, p=0.024), presence of at least one comorbidity (9 vs. 8 days, p=0.006), arrival at the hospital by ambulance vs. the patient's own transport (11 vs. 8 days, p=0.003), severe lung involvement shown on X-ray (10 vs. 8 days, p=0.030) or CT (18 vs. 10 days, p=0.045). Prolonged hospital length of stay is associated with older age (Spearman's rho=0.185, p<0.001), greater number of comorbidities (Spearman's rho=0.200, p<0.001), lower oxygen saturation on admission (Spearman's rho=- 0.294, p<0.001) and lower lymphocytes count (Spearman's rho=-0.209, p<0.001), as well as higher CRP (Spearman's rho=0.168, p<0.001), LDH (Spearman's rho=0.140, p=0.004), ferritin (Spearman's rho=0.143, p=0.004) and d-dimer (Spearman's rho=0.207, p<0.001). The multiple linear regression model found that the increase in the number of bed days of discharged from COVID-19 unit patients depends on the way the patient arrived at the Emergency Department (by ambulance instead of on their own transportation) and the presence of an accompanying oncological disease (R2=0.628, p<0.001). The hospital length of stay of patients discharged from COVID-19 intensive care unit is associated with the presence of hypertension (median 14 vs. 9 days for patients without the disease, p=0.067 - significance not reached) and at least one comorbidity (14 vs. 9 days, p=0.067 - significance not reached). The number of bed days is higher when recorded more comorbidities (Spearman's rho=0.818, p=0.004), lower oxygen saturation (Spearman's rho=-0.605, p=0.067 - significance not reached) and higher leukocytes count (Spearman's rho=0.546, p=0.102 - significance not reached). A multiple linear regression model demonstrated the hospital length of stay of patients in the COVID-19 intensive care unit as an outcome of the number of comorbidities only (R2=0.826, p=0.003). The ability to estimate and forecast quickly the number of bed-days based on a small number of variables would help reduce the burden on the healthcare system during a pandemic.
Résumé
In this article, we present a case of COVID-19 pneumonia, which arrived to us through the ER department at our hospital. At the time, this was our first contact with the disease and during the patient's stay, we came across many organizational as well as therapeutical difficulties. The RT-PCR test from the upper airways was thought to be enough to triage the patient, and as we did not possess any serological tests, was the only test we could perform. Unfortunately, as much of the literature shows, the incidence of false negative upper airway RT-PCR tests is exceedingly great. Although the reason for this is uncertain at this point in time, other diagnostic tests also fall short in terms of being able to triage the patient early on - a procedure vital in terms of patient outcome, as well as staff well-being. Radiological signs, however, can be very efficient in arousing suspicion towards such an infection, and have been shown to be extremely sensitive to early changes associated with the disease. It was indeed the chest CT, combined with the laboratory findings that lead our team to be sure of the diagnosis, despite the negative tests. At the moment of writing this article, much of the laboratory tests remain inconclusive and we have much to learn in order to be able to cope with the potential threat of this disease. What is sure, however, is that radiological identification of COVID-19 through serial chest CT scans must be made part of the baseline algorithm, so that all effort can be made to maximize what we have at out disposal. © 2020 Izdatelstvo Medicina i Fizkultura. All rights reserved.