1. IntroductionDuring the current coronavirus disease 2019 (COVID-19) pandemic, antimicrobial use increased tremendously due to the lack of proper treatment strategies [1]. Although COVID-19 is a viral disease that is untreatable by antibiotics, viral respiratory infections can progress to bacterial pneumonia, co-infection, bacterial superinfection, and other nosocomial infections requiring antibiotic administration.The overuse of antibiotics for treating COVID-19 patients was a result of (a) panic about an unknown disease, (b) similar symptoms to pneumonia, and (c) a higher death rate in communities with weaker immunity [2,3,4]. Moreover, the bacterial co-infection rate was almost 16%, and the use of antibiotics (especially broad-spectrum) increased to more than 72% during the current pandemic [3].The consumption of antibiotics during COVID-19 increased tremendously, and there are various factors contributing to the spread of AMR [5,6]. Enhanced hospital exposure is silently contributing to the emerging rate of antimicrobial resistance (AMR), which causes about 700,000 deaths per year globally [7]. The widespread use of antibiotics to control pandemics might also increase resistant organisms [8,9]. The current state of antibiotic use in COVID-19 projects approximately 10 million deaths by 2050 [9,10,11,12,13]. The evidence suggests that the COVID-19 pandemic is increasing the rate of AMR through the unnecessary use of antibiotics [13]. Therefore, it is critical to strengthen antimicrobial stewardship (AMS) and formulate policies for the use of antibiotics [12].The driving factors for the use of antibiotics include lack of proper awareness in the public, potential access and affordability to antibiotics without prescription, and use of leftover antibiotics from earlier prescriptions. While some other driving factors are insufficient training during the early phase of clinical practice, irrational prescriptions to promote a pharmaceutical company, and inadequate diagnostic process [14].The lack of new antibiotics development for the past three decades, termed as “discovery void”, is due to the scarcity of research to find new antimicrobials [12,14]. COVID-19 has also disrupted the production, delivery, and processing of antimicrobials. During the current pandemic, the demand to find a treatment has led to a shift in research resources and funds to new antivirals and vaccines instead of antibiotics [15]. Furthermore, the shortage and rerouting of medical funds during COVID-19 have affected many small labs that produce medicines and vaccines for local markets. This deficiency of narrow-spectrum antibiotics can cause an increase in AMR [16]. Clinical trials were also disrupted as the hospitals focused on COVID-19 [17]. Further, there is a need to start clinical trials of many antimicrobials. The results of these trials are expected to improve COVID-19 treatments and patient outcomes. The studies focusing on AMR have raised concern about the inflammatory effects of administered drugs [18]. Therefore, different countries have formulated many guidelines for antimicrobial use during the pandemic. However, the World Health Organization (WHO) has recommended avoiding antibiotic use for patients with mild to moderate symptoms of bacterial or COVID-19 infections [19]. According to WHO, for severe cases, only low-potency antibiotics are recommended, and in the case of aged persons, the antibiotics included in the access list of WHO (https://aware.essentialmeds.org/groups (accessed on 25 August 2021)) can only be prescribed [20]. Nonetheless, these guidelines are insufficient to limit AMR emergence. There is a strict need for evidence-based guidelines for AMS during and/or post-COVID-19 pandemic. There is a need to analyze the trends for the pandemic’s spread and the complete details of antibiotics used globally since the COVID-19 outbreak.

Here, we performed a systematic analysis to assess changes in antibiotic use during the COVID-19 era and how these changes might impact AMR.

3. Discussion

The extensive use of antibiotics has increased since the COVID-19 outbreak, increasing concerns about AMR. This study includes only the data of hospitalized patients because very little data are available about those confined to their homes with mild or moderate symptoms. This is probably because most research studies are performed on patients reporting to medical facilities.

We found that >78% of COVID-19 patients were recommended to use antibiotics (Figure 2). The broad-spectrum antibiotic azithromycin was the most frequently prescribed, followed by ceftriaxone, moxifloxacin, meropenem, and tazobactam. The excessive use of broad-spectrum antibiotics without proper clinical justifications by healthcare persons has raised concerns about AMR amplification [21].It has been previously reported that azithromycin was the most common antimicrobial agent used while treating COVID-19 [22]. Azithromycin is very efficient in treating pneumonia, but there is no proof of its effect on viruses. There are studies that clearly show that excessive use of azithromycin can result in antimicrobial resistance, which indicates that overuse or misuse of this antibiotic during COVID-19 can contribute to AMR [23,24]. The WHO also reported the excess use of azithromycin for COVID-19 treatment even without approval [25]. It is classified as a critically important antimicrobial used in humans [26]. Persistent use of this antimicrobial can result in AMR, causing a serious threat to survival in severe infection cases. Third-generation cephalosporins such as ceftriaxone have long been mostly used in intensive care units (ICUs). As the consumption of other antibiotics changed during the pandemic, cephalosporins were also being repurposed [27]. Karami et al. provided clear evidence of cephalosporin use in more than 200 of 556 cases [28].Doxycycline is another drug that could help treat COVID-19, which is used because of its antiviral and anti-inflammatory properties [29]. Many broad-spectrum antibiotics, such as fluoroquinolones and cephalosporins, were reportedly used in more than 74% of patients. If this use of broad-spectrum antibiotics is not stopped, it could lead to AMR. This could lead to very few remaining options for treating infections, and these options could become unaffordable for underdeveloped countries.Increasing telehealth services also caused excessive antibiotic use due to the unavailability of the proper diagnostic channel and determining the nature of the illness [30]. The best treatment options for COVID-19 were unknown, so experimental treatments played a large role in inappropriate drug prescriptions during this pandemic period [17]. Due to this vagueness in the COVID-19 era, healthcare workers prescribed antibiotics based on the assumption that the drug’s potential threat would be negligible compared to its benefit [31]. Previous studies also mentioned increased antimicrobial use due to self-medication [32,33,34]. This was more common in areas where antimicrobials could be easily accessed without any prescription [16,35]. One study reported that to avoid visits to any healthcare facility, about 20% of Iranians used self-prescribed medicines for their sickness [36]. We found that the recommendation rate of antibiotics does not change with the severity of the disease (Figure 2). Both mild or moderate and severe or critical illness patient groups were given antibiotics, although the severe cases had a greater chance of suffering from a secondary infection. The high rate of antibiotic application in mild cases is also very alarming. The results show that the mortality rate was higher in cases where all patients used antibiotics than in cases in which most patients were not given antibiotics. Length of hospital stay (LOS) was higher in the patients’ group, where not all, but the majority, were given antibiotics (Figure 3). In our analysis, 51.3% of the patients showed secondary infections, and 57.5% were in critical condition (Table 1). We also found that the co-infection rate was higher in patients with severe COVID-19 symptoms, and the mortality rate was greater in patients with some co-infection or secondary infection (Figure 4). Although the rate of antibiotic prescription is enormous, secondary infections were only reported in 14.3% of cases, along with 3.5% of reported cases of co-infections [20]. The bacterial co-infection of COVID-19 patients has been reported in many studies worldwide [37,38,39,40,41]. The co-infection rate was almost 28% in Europe [42]. Patients with mild or medium symptoms were not reported for co-infection because these patients were not checked or tested for infection [43,44,45]. Moreover, in many cases, when the specimen was taken, the co-infections were reported to be more associated with hospital-acquired than community-acquired infections. Our analysis shows that the number of antimicrobial types increased only slightly in 2020 and 2021 compared to 2019, but the number of patients using antimicrobials increased tremendously (a four-fold increase) (Figure 5). The total number of admitted patients also increased from 2019 to 2021. This indicates a positive relationship between the number of COVID-19 patients and the antimicrobials consumption, contributing to AMR. A higher rate of AMR could be predicted in low- and middle-income countries because of a lack of awareness and stewardship programs, poor lab facilities, and a lack of proper rules for accessing antibiotics without prescription [46]. COVID-19 can be more easily spread to areas that are more populated and lack proper hygiene facilities. In low- and middle-income countries, allocating resources to COVID-19 is very difficult as their healthcare facilities already lack proper funds, which is an additional burden to their healthcare systems [32].In many studies, classes of antibiotics are mentioned, but information related to the use of antibiotics according to disease severity was not available. The variation in the number of research studies across regions might have affected our results, as there is a huge difference in the number of patients and local regulations for COVID-19 [47]. Moreover, the selection biases (not all the studies included in this systematic review are directly discussing AMR and COVID-19) are there. In this paper, we mostly focused on selecting the papers that provide evidence of antibiotic use during COVID-19, as the data related to antibiotics use can help in providing a clear understanding of the rate of AMR prevalence at this time. Healthcare professionals were also under pressure during the COVID-19 pandemic. In small hospitals where healthcare facilities are insufficient, they must try every possible option to save the life of patients. The diagnosis of secondary infections is very costly, and most hospitals lack this facility, leading doctors to prescribe antibiotics even when they are not needed. This needs to be managed as the excessive use of antimicrobials raises the threat of AMR. Antimicrobial Stewardship (AMS)

Our analysis demonstrates that COVID-19 has crucial implications for AMR. The overuse or misuse of antibiotics to control COVID-19 symptoms, even without any co-infection, has worsened the situation. The world health organization has clear guidelines in this regard; patients with mild symptoms should not be prescribed antibiotics. The antibiotics should only be prescribed when there is clear evidence of bacterial co-infection. Better diagnostics are required to identify patients with secondary infections to avoid misuse of antimicrobials. AMS programs can assist in properly using antimicrobials by reviewing every prescribed medication. The data to date show that the use of antimicrobials is much higher than needed. For real-time data review and dissemination, additional efforts are needed to improve AMS. These efforts will not only help control COVID-19 but will have a major role in controlling the future pandemic of AMR.

4. Materials and MethodsThe study was conducted to identify and analyze the research studies reporting the use of antimicrobials (especially azithromycin, doxycycline, clarithromycin, ceftriaxone, erythromycin, amoxicillin, amoxicillin-clavulanic acid, ampicillin, gentamicin, benzylpenicillin) for treating COVID-19. The data were reported using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for systematic review, and its protocol is registered in the PROSPERO register [48,49]. In order to maximize the authenticity of our findings, we tried to include as many research studies as possible in this analysis. All data were from hospitalized COVID-19 patients, and the final selection was based on RT-PCR-confirmed COVID-19 cases. 4.1. Strategy for Data Search

The study included all papers from several scientific databases such as Scopus, Web of Science, and PubMed-associated peer-reviewed journals published since the COVID outbreak (December 2019 till December 2021). The keywords and search terms used were “Antimicrobial resistance” and “coronavirus” or “COVID-19” and “Antimicrobial resistance” and “Antimicrobial stewardship” or “Antibiotic resistance” and “COVID-19”.

4.2. Enclosure and Elimination Criteria for Research Studies

We included all the studies: (1) related to COVID-19 patients from communities and hospitals; (2) where antibiotics were prescribed during treatment; and (3) published in English (or in press). We selected cohort studies, cross-sectional studies, case-control studies, randomized control studies, and descriptive and observational research studies related to the use of antimicrobials in COVID-19 patients. We included those studies reporting the patients and use of antibiotics without any discrimination in gender, age, color, country, or community. We also considered the studies that reported antibiotics usage without specifying their types or treatment outcomes. We excluded those research studies that overlapped (duplicate data), contained unreliable data (short reports containing no proper results), were published in the form of editorials and notes, and studies related to engineering and earth sciences. Our analysis did not consider the studies related to animal experimentation, molecular mechanism, drug modeling, and other aspects of COVID-19, except for the use of antibiotics/antimicrobials.

4.3. Data Extraction

After cross-checking for the study accuracy and duplicates, data extraction was carried out based on the year of study, type of article, design or idea of the study, area/country of study, the sternness of COVID-19, rate of bacterial co-infections (infection acquired with first 24 to 48 h of hospitalization) and secondary infections (infection acquired after 24 to 48 h of hospitalization), prescribed antimicrobials, and the number of patients consuming those antimicrobials. The extraction of data from selected research studies comprised of details of publication, region of study, number of reported patients, type of study (e.g., case study, cohort study, descriptive study), condition of patients (mild, severe, or critical), rate of antimicrobial usage and recommendation, the scenarios of prescription, time of hospital stay, and mortality rate. More than half of the studies reported disease severity as mild, moderate, critical, and severe, while the others used just mild, moderate, and severe. We categorized disease severity into two major groups: mild or moderate and severe or critical.

4.4. Data Synthesis and Analysis

We analyzed all the selected studies where details about medicine prescriptions were available. We also investigated antimicrobials and antibiotics use according to the severity of illness, types that were used most often to treat COVID-19, health effects related to the use of antibiotics, bacterial co-infections, and related outputs, and trends of antimicrobial usage at the time of COVID-19. We also explored the length of hospital stay (LOS), rate of discharge, and mortality (patients still in the hospital at the time of publication were excluded from this calculation).