By observing the cohort of patients, it emerged that males were more affected by COVID-19 compared to females, regardless of the clinical outcome and the pandemic phase. This data confirms previous studies, in which a higher percentage of males with COVID-19 were reported [22]. Different causes, for example the expression of ACE2 receptor, have been considered as an indicator of gender susceptibility [23].
Based on the demographic data, the mean age of the non-surviving patients was significantly higher than the surviving ones, as expected. If only deceased patients are considered, it is interesting to note that the age of the patients in the first phase was higher, probably because SARS-CoV-2 infection hit the oldest and more susceptible individuals in the population. Other risk factors were also reported, and the percentage of smokers was twice as high in deceased patients compared to the transferred patients, suggesting cigarette smoke as a negative predicting factor of the COVID-19 prognosis [24].
No differences resulted from BMI and alcohol consumption analysis. Several co-morbidities, including hypertension, heart failure, atrial fibrillation, asthma, chronic obstructive pulmonary disease and chronic renal failure were more frequent in deceased patients [25]. The deceased patients appeared more susceptible to contracting secondary infections and had a reduced ability to compensate for the alterations induced by them [26]. Regarding the onset of symptoms, our data show that only dyspnoea, hyposmia/hypogeusia, and diarrhea have a higher incidence among deceased patients.
Blood chemistry parameters
Covid-19, and possibly associated secondary infections, caused various laboratory alterations, with a different trend between the two groups analyzed. In deceased patients, there was a statistically significant increase in the indices of inflammation, creatinine, sodium, and potassium, and a significant reduction in hemoglobin and lymphocytes, probably due to the organism's failure to counteract infections and consequent organ damage [27]. The same laboratory parameters in surviving patients showed a significant increase in lymphocytes, platelets, sodium, and potassium plus a reduction in the indices of inflammation, hemoglobin, and markers of liver damage [28]. These changes are consistent with a probable resolution of the infections and, therefore, improve the clinical outcome.
In this study, the changes in our patients' laboratory test results over time and the differences between the three pandemic waves were also analyzed. Regarding the group of deceased patients, all the considered parameters were stable in the first and third waves. In contrast, in the second wave, leukocytes, procalcitonin, C reactive protein, D-dimer, creatinine, LDH, AST, and ALT increased considerably, particularly from T1 to death, while lymphocytes were reduced. These differences over time are probably linked to the fact that in September - December 2020, the patients admitted to the Nuoro ICU were more affected by other bacterial/fungal infections which may have contributed to the poor prognosis of these patients. Regarding the transferred patients, the trend over time of the laboratory results in the three waves was rather stable in the three periods analyzed with minimal variations. In particular, a reduction in the inflammation indices was observed in the first wave, probably because, in our region, this was the phase of the pandemic with fewer COVID-19 cases and fewer deaths, likely related to the geographical position and the severe restrictive measures adopted by the government (national lockdown).
Based on the comparison of the laboratory parameters between the deceased and transferred patients at the admission time (T0), we didn’t find any significant result, and this evidenced the lack of the predictive role of them in terms of prognosis and outcome of the COVID-19 disease.
Microbiological data
Another crucial aspect of this study was to focus on secondary infections, which comprise the group of all patients with a positive culture (urine, bronchoaspirate, blood culture, and CVC) during hospitalization in the ICU. From our analysis, it emerges that among the 156 patients, approximately 44% presented positive cultures, and of these, 80% died. By analyzing the data from the literature, the percentage of COVID-19 patients with secondary infections is highly variable, ranging between 14% and 100%, depending on the different inclusion criteria used [8, 15, 29].
Differences in the population, specimen source, and pathogens of interest are likely responsible for the wide variations reported [30]. We observed a high percentage of positive samples, and it is probably because in our study, we considered all the positive cultures occurring after 7 days of hospitalization, including blood and urine cultures, bronchoaspirate and CVC cultures. Another critical aspect to consider is the logistical conditions in which the healthcare personnel had to work, especially during the second wave, as there was a rapid and sudden increase in COVID-19 cases and ICU admissions. The higher incidence of catheter-related bloodstream infection could be explained by the heavy pressure put on the ICU by the COVID-19 outbreak; in fact, during the peak, the ICU capacity of the Nuoro hospital had to be increased by 300% to accommodate all patients requiring critical care. Interestingly, from the microbiological analysis, it emerged that deceased patients showed a higher percentage of positive samples than the transferred patients considering blood cultures and bronchoaspirate. On the other hand, a higher percentage of positive samples in urine cultures and CVC cultures was observed in the group of transferred patients suggesting that it is correlated with the longer period of hospitalization. Indeed, the average ICU stay for the transferred patients was longer than for deceased patients (31 vs 17 days, respectively), explaining the greater occurrence of nosocomial infections in the urine and CVC cultures of the first group of patients.
Our attention focused on the microorganisms likely responsible for the secondary infections in the two groups of patients. Staphylococcus aureus, Klebsiella pneumonia, Candida spp, Enterococcus faecalis, Pseudomonas aeruginosa and above all CoNS were mainly detected in both groups. These findings are in line with other published studies [15]. During the hospitalization of our patients, Klebsiella pneumoniae isolates developed acquired resistance mechanisms, such as ESBL (Extended-Spectrum β-Lactamase) and carbapenem-resistance, determining a difficult challenge for the choice of the antibiotic treatment, especially during the second wave.
Our results are overall consistent with what is already reported in the literature. Microorganisms mainly involved in the COVID-19 secondary infections were Mycoplasma spp., Haemophilus influenzae, P. aeruginosa and S. aureus, with a discrete variability in the species most represented and widespread in the various territories and hospitals [12, 31].
Based on our results, it seems clear that secondary infections may have played a critical role in the negative outcome of the patients affected by COVID-19. Secondary infections may have acted as mutually reinforcing factors to promote progression to severe and fatal disease. Indeed, severe COVID-19 caused multiple damage [32], which may favor rapid bacterial growth; on the other hand, bacterial virulence factors can alter the immune responses, resulting in a rebound of SARS-CoV-2 infection and disease progression, leading to higher mortality in severe and critical patients [33, 34].
Finally, the analysis of our data revealed that the most used antibiotics were Azithromycin (60%), Piperacillin-Tazobactam (40%), and Vancomycin (20%). Such a high percentage of Azithromycin could be attributed to the fact that, in many cases, it was already administered to the patients before hospitalization or in the infectious disease divisions to patients with symptomatic COVID-19. However, despite Azithromycin has been one of the few drugs used in COVID-19 patients during the start of the pandemic, there is currently no evidence to justify the widespread use of this antibiotic [35]. As for Piperacillin-Tazobactam, it has been used in many patients as a broad-spectrum empirical therapy, pending the microbiological and antibiogram results. Finally, Vancomycin was also used in a high percentage of patients to treat infections caused by Gram-positive bacteria resistant to other classes of antibiotics.