Laboratory parameters associated with prolonged hospital length of stay in COVID-19 patients in Johannesburg, South Africa

Abstract

BACKGROUND. Coronavirus disease (COVID-19) has imposed unprecedented stressors on South Africa (SA)’s healthcare system. Superimposed on the country’s quadruple burden of disease, pandemic-related care further exposes existing inequities. Some of these inequities are specific to hospital-based inpatient services, such as the geographical maldistribution of hospital beds, lack of oxygen supplies and assisted ventilation, and scarcity of trained healthcare workers. Certain high-risk groups, such as individuals with cardiometabolic comorbidity, are likely to develop severe COVID-19 disease requiring hospitalisation with potential for a prolonged length of stay (LoS). It may be helpful for health authorities to identify those at risk for prolonged LoS to facilitate appropriate health systems planning. OBJECTIVES. To identify hospital admission laboratory parameters associated with a hospital stay >14 days in patients with COVID-19 pneumonia. METHODS. A retrospective observational study design was used. Laboratory data were obtained from an SA private laboratory for 642 inpatients with suspected or confirmed COVID-19 pneumonia, comprising 7 months of admission laboratory data from six private hospitals in Johannesburg, Gauteng Province. RESULTS. Of 642 hospital admissions for pneumonia, 497 were confirmed to have COVID-19 infection (reverse transcription-polymerase chain reaction test positive). In the COVID-19-positive group, hospital LoS was prolonged in 35.4% of admissions. Univariate analysis demonstrated an association with the following risk factors for prolonged LoS: older age; male sex; high serum creatinine, sodium (Na), chloride, potassium and urea levels and low estimated glomerular filtration rate; raised white blood cell count, lymphopenia, neutrophilia and an elevated neutrophil-to-lymphocyte ratio (NLR); and elevated levels of D-dimers, interleukin-6 (IL-6), and procalcitonin (PCT). The strongest univariate associations (relative risk (RR) ≥2.0) with a hospital stay >14 days were high Na levels, NRL >18, high PCT levels and IL-6 >40 pg/mL. On multivariable analysis, the following factors remained significantly associated with prolonged LoS: older age (RR 1.015 per year of age; 95% confidence interval (CI) 1.005 - 1.024); hypernatraemia (RR 1.80; 95% CI 1.25 - 2.60); hyperkalaemia (RR 1.61; 95% CI 1.18 - 2.20); and neutrophilia (RR 1.47; 95% CI 1.15 - 1.88). CONCLUSIONS. COVID-19 pandemic preparedness requires hospital-based inpatient care to be prioritised in resource-limited settings, and availability of beds and prompt admissions are essential to ensure good clinical outcomes. In this study of COVID-19 patients admitted with pneumonia, multivariable analysis showed older age, hypernatraemia, hyperkalaemia and neutrophilia to be associated with LoS >14 days. This may assist with healthcare systems planning.