Search for a command to run...
The aim of the present thesis is to deepen the knowledge of severe and critical COVID-19 infection - its management and outcomes among ICU patients in Sweden. This is achieved through retrospective studies of ICU patients with COVID-19 from several Swedish hospitals focusing on their management and outcomes (study I, III, IV). In addition, a prospective follow-up study was conducted on patients with post covid condition (PCC) (study II). In the first paper (study I), all patients with respiratory failure due to COVID-19 who were admitted to an ICU in region Östergötland during the initial pandemic wave were included (n=100). The median age was 63 years, and the 60-day mortality rate was 22% across the entire pandemic wave. Divided into three consecutive tertiles, the 60-day mortality decreased from 33% in the first tertile to 15% and 18% in the subsequent two. Ninety-one percent had at least moderate acute respiratory distress syndrome (ARDS) and 88% required invasive mechanical ventilation (IMV). During the wave, the use of thromboprophylaxis increased, the steep rise in ICU admissions subsided, and ICU resources expanded. At four-month follow-up, 63% of survivors reported a decline in general health compared with their health status prior to SARS-CoV-2 infection. In Study II, all patients with PCC at a clinical follow-up four months after hospital discharge in region Östergötland were included and interviewed two years after initial infection. Of 181 eligible patients, 165 participated in the study. The majority (84%) reported lingering problems affecting everyday life. Nevertheless, improvements were observed in both prevalence and severity of several symptoms and functional limitations compared with four months post-discharge. The most reported symptoms were cognitive, sensorimotor, and fatigue related. Comparison between ICU-treated and non-ICU-treated patients revealed no significant difference at 24-months. In study III, factors associated with ventilator-associated lower respiratory tract infection (VA-LRTI) in COVID-19 were explored. All patients with respiratory failure requiring IMV who were admitted to an ICU in the southeast healthcare region of Sweden were included (n=536). Overall, 28.5% developed VA-LRTI, corresponding to an incidence rate of 20.8 first VA-LTRI episodes per 1000 IMV days. The incidence of VA-LRTI increased from 14.5 per 1000 days of IMV days during the first wave to 24.8 per 1000 IMV days during the subsequent two ways. Patients who developed VA-LRTI had fewer ventilator-free days, received corticosteroids more frequently, and were more often ventilated in prone position. Most detected pathogens were Enterobacteriaceae (38.9%) and Staphylococcus aureus (22.8%). Logistic regression analysis revealed significantly increased adjusted odds ratio (aOR) for first VA-LRTI for corticosteroid treatment (aOR 2.64 [95% confidence interval [CI]] [1.31–5.74]), antibiotics at intubation (aOR 2.01 95% CI [1.14–3.66]), and days of IMV (aOR 1.05 per day of IMV, 95% CI [1.03–1.07]). In the final paper (study IV), mortality disparities among patients with COVID-19 admitted to ICUs across seven Swedish hospitals were investigated using survival analysis. All patients admitted to one of the participating ICUs with respiratory failure due to COVID-19 were included (n=747). Across the cohort, 90-day mortality varied substantially between hospitals, ranging from 8.5% to 30%. In the final cox proportional hazards model adjusted for baseline covariates, pandemic wave and with random intercept for healthcare county, the adjusted hazard ratios (aHR) for 90‑day mortality by hospital spanned from: 2.38 to 5.06, using the hospital with the lowest mortality as reference. The results remained robust after sensitivity analysis, including complete case analysis, calculation of e-values, assessment of multicollinearity, and testing of the proportional hazards assumption. In conclusion, initial high mortality of ICU-treated COVID-19 patients in region Östergötland, quickly declined during the first pandemic wave, paralleling increased ICU resources and expanded use of thromboprophylaxis. Moreover, many ICU survivors experienced reduced general health at four-month follow-up. Many survivors developed PCC and reported lingering symptoms affecting their everyday life two years after initial infection, but with significant improvement compared to the initial follow-up. Additionally, the VALRTI incidence in southeast healthcare region of Sweden was low compared with previously published data, although it increased across the pandemic waves - possibly related to expanded use of corticosteroid treatment. Finally, during the pandemic, the initial ICU to which a patient was admitted correlated with ICU mortality. These results contribute to the ongoing discussion regarding healthcare equity in Sweden, suggesting that geographical inequities were likely present at least during the pandemic.