Discussion

Among the study population, 57.8% male patients were identified in the prepandemic period, while 58.4% male patients were identified in the postpandemic. Additionally, the age group over 60 years old predominated in both periods. This observation aligns with the report of the Bahçe et al.’s study, which compared the profiles of bacterial isolates from tracheal aspirates between the prepandemic (April/2019 to March/2020) and postpandemic (April/2020 to March/2021) periods in Turkey. ^{Reference Bahçe, Acer and Özüdoğru12} Their findings also demonstrated a predominance of male patients in both periods, with an average age range similar to that reported in our study and with no statistical difference between the periods (P = .710). Male patients predominated in both periods, reflecting the demographic profile of the ICU population.

Among the microorganisms analyzed, Gram-negative bacteria were the most frequently isolated, accounting for 301 isolates, which constituted 90.3% of the total samples (including Gram positive, Gram negative, and fungi). P. aeruginosa, K. pneumoniae, and A. baumannii were prevalent in both periods. These same GNB were isolated and identified in other studies, but with different frequencies. ^{Reference Bahçe, Acer and Özüdoğru12–Reference Warda, Molham and Salem15} They have been recognized as the primary agents of lower respiratory tract HAIs in ICU patients even before the COVID-19 pandemic. ^{Reference Kelliher and Kirton16}

Bacterial isolates classified as MDR, XDR, or (PDR) accounted for 45% of HAI/LRTI caused by bacteria in P1 and 51.6% in P2. A single-center study conducted in Italy reported a higher proportion of postpandemic resistance, with 64.5% of MDRO isolates identified in patients with ventilator-associated pneumonia, albeit based on a smaller sample of 48 patients with infections. ^{Reference Temperoni, Caiazzo and Barchiesi3} In a study of more than 8,000 healthcare-related infections in Brazil, 29.7% were caused by MDROs. One of the possible explanations for this difference between the studies is the number of participants and the locations of the studies. Hence, the local and socioeconomic reality of the countries most affected by the increase in MDR bacteria must be considered. ^{Reference Polly, Almeida, Lennon, Cortês, Costa and Guimarães17}

Several factors have contributed to this scenario. Initial reports from China indicated that over 50% of deaths among COVID-19 patients were associated with secondary bacterial infection. ^{Reference Zhou, Yu and Du18} Kanungo et al reported antibiotic prescription in 60%–70% of patients with COVID-19. ^{Reference Kanungo19} Conversely, Rawson et al’s (2020) systematic review identified only 8% of bacterial or fungal co-infection in patients with COVID-19, out of 800 individuals. ^{Reference Rawson, Moore and Zhu20} However, 72% received empirical antibiotic therapy. In India, a multicenter study involving 10 hospitals found that 47% of patients received carbapenems. ^{Reference Vijay, Bansal and Rao21} During this time, the WHO warned against the indiscriminate use of azithromycin and hydroxychloroquine, emphasizing that there was no recommendation for their use outside of controlled clinical trials. ²² Shockingly, data indicate that in some areas, 48.53% of patients used antibiotics without medical advice. ^{Reference Sulayyim, Ismail, Hamid and Ghafar23} Detailed individual antibiotic exposure data and stratified analyses by COVID-19 status were not available for all patients, representing a limitation of this retrospective surveillance-based study.

Other factors may explain the increased prevalence of MDROs: the high intensity of care required by critically ill patients during the COVID-19 pandemic; prolonged patient contact during procedures such as pronation, requiring four to five healthcare professionals (physiotherapist, nursing technician, nurse, and doctor); and the presence of inexperienced ICU professionals. ^{Reference Tiri, Sensi and Marsiliani24} Most MDRO bacterial infections have developed in critically ill patients, resulting in prolonged hospitalization. ^{Reference Rawson, Moore and Zhu20,Reference Lai, Wang and Hsueh25–Reference Wang, Hu and Hu28}

The widespread use of antibiotics prophylactically or at an early stage has also contributed to bacterial resistance worldwide, with more than half of patients with COVID-19 receiving antibiotics. ^{Reference Lai, Wang and Hsueh25,Reference López-Jácome, Fernández-Rodríguez and Franco-Cendejas29,Reference Mustafa, Tolaj, Baftiu and Fejza30} This impact is particularly significant in underdeveloped or developing countries, where infection control programs and the rational use of antimicrobials have not yet been implemented. ³¹

The excessive use of antibiotics during the COVID-19 pandemic can be attributed to several factors: similarity of the picture of SARS-CoV-2 infection with community-acquired pneumonia (cough, fever, and radiological infiltrate); anxiety and uncertainty regarding the severity of the disease and its potential complications; the lack of effective medications; and the possibility of co-infection or secondary infection as a justification for administering antibiotics to patients with worsening clinical conditions. ^{Reference Lai, Chen, Ko and Hsueh32} Features such as bilateral ground-glass opacities on imaging, lymphopenia, and the absence of focal lobar consolidation are more characteristic of COVID-19 pneumonia compared with typical community-acquired bacterial pneumonia.

K. pneumoniae has emerged as a significant pathogen in healthcare-related infections in the postpandemic period of COVID-19. ^{Reference Warda, Molham and Salem15,Reference Sulayyim, Ismail, Hamid and Ghafar23,Reference López-Jácome, Fernández-Rodríguez and Franco-Cendejas29,Reference Bazaid, Barnawi and Qanash33} Resistance is primarily mediated by the production of β-lactamases and other antibiotic-inactivating enzymes. ^{Reference Tavares34} In this study, this species showed the highest level of resistance, with isolates classified as resistant to all antibiotic classes tested (PDR), a finding consistent with a study in Saudi Arabia. ^{Reference Bazaid, Barnawi and Qanash33}

In the P. aeruginosa isolates, the highest frequency of bacteria causing HAI/LTRI was quantified in the two study periods, with a significant increase in bacteria classified as MDR and XDR. Data from the United States indicated a 32% increase in MDR P. aeruginosa during the first peak of the COVID-19 pandemic. ³⁵ The rate of detection of resistance genes is low in this species, given the predominance of other resistance mechanisms, but with a tendency toward an increase in bla _NDM. ^{Reference Kiffer, Rezende and Costa-Nobre36} Despite the absence of carbapenemases in the postpandemic period, resistance rates to carbapenems, quinolones, and fourth-generation cephalosporins remained close to over 50%, consistent with findings reported by Bahçe et al. ^{Reference Bahçe, Acer and Özüdoğru12}

A. baumannii isolates showed an increase in their resistance profile between the periods, with a significant impact on antimicrobial resistance to six antibiotics. These findings align with those of Bazaid et al, who found resistance to all the tested antibiotic classes in 65 ICU patients with infections involving different anatomic sites. ^{Reference Bazaid, Barnawi and Qanash33} Notably, their study revealed that 100% A. baumannii isolates showed resistance to carbapenems, 87% to amikacin, and 53% to colistin. Although specific carbapenemase profiles were not identified in these bacteria, other studies showed that _blaOXA-23, _blaOXA-143, and _blaOXA-58 were the most prevalent in A. baumannii. ^{Reference Kiffer, Rezende and Costa-Nobre36,Reference Pillonetto, Jordão and Andraus37}

Infection control programs have had major barriers to maintaining or implementing the necessary measures to tackle MDR bacteria during the COVID-19 pandemic. ^{Reference Abubakar, Al-Anazi, Alanazi and Rodríguez-Baño38} Gram-negative bacteria are the main agents of healthcare-related infections, with an impact on treatment, contact precautions, and cause of outbreaks in hospital units, and with an increase in quantity and resistance profile in the postpandemic period of COVID-19. ^{Reference Gandra, Alvarez-Uria and Stwalley39} Carbapenemases have increased the resistance potential of these bacteria and are spreading in various regions of the country. ^{Reference Kiffer, Rezende and Costa-Nobre36} The impact of resistance goes beyond the ICU setting; long-stay institutions have also experienced a shift in their bacterial resistance profile. ^{Reference Tomczyk, Taylor and Brown40}

Healthcare facilities must develop strategies to address this scenario of MDR bacteria in LRTIs and implement quality of care measures such as mechanical ventilation bundles, initiatives to reinforce the correct technique and adherence to hand hygiene, and proper cleaning of units and devices, which consequently has an impact on healthcare-related infections caused by MDROs. ^{Reference Cole and Barnard41}

To combat antimicrobial resistance effectively, laboratory and hospital monitoring needs to be expanded and strengthened, as it is through the data obtained that assertive actions can be drawn up at local, state, and federal levels. Flows, protocols, and trainings in the detection of these resistant bacteria are extremely important to improve the recovery of samples, the investigation of outbreaks, and the notification of critical priority microorganisms in both public and private institutions. Accurate differentiation between colonization and true infection is central to antimicrobial stewardship, as inappropriate treatment of colonization can unnecessarily increase selective pressure for antimicrobial resistance.

The change in the profile of antibiotic resistance directly affects the chances of a cure and the quality of hospital care, given the increase in serious cases, especially respiratory infections, in ICUs.