Autore: Antonio Sciurti, Dipartimento di Sanità Pubblica e Malattie Infettive Sapienza Università di Roma
antonio.sciurti@uniroma1.it
Mariateresa Ceparano, Dipartimento di Sanità Pubblica e Malattie Infettive Sapienza Università di Roma; Claudia Isonne, Dipartimento di Sanità Pubblica e Malattie Infettive Sapienza Università di Roma; Giuseppe Migliara, Dipartimento di Sanità Pubblica e Malattie Infettive Sapienza Università di Roma; Valentina Baccolini, Dipartimento di Sanità Pubblica e Malattie Infettive Sapienza Università di Roma; Jessica Iera, Dipartimento di Sanità Pubblica e Malattie Infettive Sapienza Università di Roma; Alessandro Territo, Dipartimento di Sanità Pubblica e Malattie Infettive Sapienza Università di Roma; Luca Tentarelli, Dipartimento di Sanità Pubblica e Malattie Infettive Sapienza Università di Roma; Paolo Franzelletti, Dipartimento di Sanità Pubblica e Malattie Infettive Sapienza Università di Roma; Carolina Marzuillo, Dipartimento di Sanità Pubblica e Malattie Infettive Sapienza Università di Roma; Paolo Villari, Dipartimento di Sanità Pubblica e Malattie Infettive Sapienza Università di Roma;
Categoria Primaria: Epidemiologia clinica
Categoria Secondaria: Malattie trasmissibili
Introduction. The shortage of critical care beds was one of the major challenges healthcare systems had to face worldwide in the COVID-19 pandemic. In this context, to manage the increasing number of COVID-19 patients, the Policlinico Umberto I in Rome, established a 18-bed temporary intensive care unit (ICU) from 1 March 2021 to 15 July 2023.
Objectives. The study aim is to describe the Healthcare-Associated Infections (HAIs) incidence and risk factors in the temporary ICU unit during the phases of the COVID-19 pandemic.
Methods. Patients were categorized by admission dates during dominant SARS-CoV-2 variant phases: Alpha (1 Mar 2021–18 Jun 2021), Delta (19 Jun 2021–31 Dec 2021), Omicron BA.1 (1 Jan 2022–12 Mar 2022), Omicron BA.2 (13 Mar 2022–6 Jun 2022), Omicron BA.4 (7 Jun 2022–12 Feb 2023), and Omicron XBB (13 Feb 2023–15 Jul 2023). HAI incidence rates and mortality per 1000 patient-days were calculated. A multivariable Fine-Gray model, with death as a competing event, was used to compute sub-distribution hazard ratios (SHRs) with 95% confidence intervals and identify predictors for first HAI.
Results. 355 patients with a ≥ 48 hours a length of stay were admitted. Mean patients’ age was 67.0 ± 13.8 years and 66.8% were male. Overall, 97 patients experienced HAIs (27.3%), and 176 (49.6%) died. Higher rates of first HAI per 1000 patient-days were observed in the Alpha (30.8, 95% CI: 17.6–44.0), Delta (36.0, 95% CI: 21.0–51.1), and Omicron BA.1 (27.9, 95% CI: 14.6–41.2) phases compared to the Omicron BA.2 (15.3, 95% CI: 5.3–25.3), Omicron BA.5 (12.7, 95% CI: 7.0–18.3), and Omicron XBB (16.4, 95% CI: 5.7–27.1) phases. A higher mortality rate per 1000 patient-days was found in Delta (44.4, 95% CI: 29.7–59.1), and Omicron BA.1 (38.8, 95% CI: 25.3–52.2) phases compared to the Alpha (20.4, 95% CI: 11.5–29.4), Omicron BA.2 (33.4, 95% CI: 19.5–47.4), Omicron BA.5 (30.6, 95% CI: 22.4–38.8), and Omicron XBB (17.9, 95% CI: 8.2–27.6) phases. Among the isolated microorganisms (N=165), A. baumannii (N=40) was most frequently responsible for HAIs, followed by K. pneumoniae (N=30). According to multivariable analysis, compared to the Alpha phase, admission during the Omicron BA.2, BA.5, and XBB phases was associated with a lower risk of HAIs (SHR: 0.30, 95% CI: 0.13–0.70, SHR: 0.22, 95% CI: 0.10–0.48, SHR: 0.32, 95% CI: 0.14–0.75, respectively), but not in the Delta and Omicron BA.1 phases (SHR: 0.67, 95% CI: 0.32–1.42, SHR: 0.42, 95% CI: 0.17–1.01, respectively). Intubation at admission was associated with a higher risk of HAIs (SHR: 1.72, 95% CI: 1.05–2.82). Male patients showed a higher HAI risk (SHR: 4.50, 95% CI: 1.87–10.83), but this effect decreased over time.
Conclusions. A higher risk of HAI was found during the first phases of the emergency. Evolving clinical practices in managing COVID-19 patients, the vaccination campaign and the different SARS-CoV-2 variants may have influenced HAI rates.