We found that the incidence of FN was significantly reduced in post COVID 19 era and that it was the only significant clinical risk-reducing factor. Since personal quarantine guidelines, including social distancing, masking, hand hygiene, were strongly recommended in South Korea after COVID 19 outbreak, our study suggest that standard precaution in medical practice are effective strategy to prevent FN. A similar pattern was also observed in seasonal respiratory virus infection rates (11).
Cytotoxic chemotherapy induced myelosuppression disrupts mucosal integrity related to FN development. Therefore, although patients and tumor factors also affect NF occurrence and pathogenesis, seeding of gut normal flora in patients is considered the most important cause of FN. Approximately 80 % of documented infectious organism are considered of an endogenous origin (12). For this reason, several guidelines for preventive strategies using CSFs and antibiotics are emphasized, and non-pharmacological measure, such as standard precaution, are usually less emphasized. Although the 2011 IDSA guidelines for neutropenic patients with cancer mentioned that hand hygiene is the most effective preventive measure, they do not provide direct evidence for FN, and only provide indirect evidence for hospital-acquired infection prevention (5, 8, 13-15). Because patients with cancer are a very heterogeneous disease group, the etiology of FN inevitably differs among patients, and reportedly viral agent are also important etiology of FN (16-18). In that respect, the preventive strategy for FN also needs to be applied differently depending on the patient's characteristics and expected etiology; however, the current practice is considered to too simplistic and is focused on only a specific group. Our data and some recent studies have shown that the incidence of FN, especially upper respiratory infection, significantly decrease after COVID 19 outbreak (19, 20). In our study, the pattern of our relatively lower risk patients was different from the general known epidemiological pattern of FN (Unexplained fever, 87.5%; clinically documented infection, 6.3%; and microbiologically documented infection, 6.3% respectively). These results imply that environmental infectious source also plays an important role of FN occurrence and non-pharmacological intervention, such as standard precautions, may be more meaningful in some groups. This also suggests that preventive guideline of FN should be more individualized according to patient’s clinical factors. For example, prophylactic antibiotics and CFSs should be strongly considered in higher risk patients, and non-pharmacological intervention should be relatively emphasized in lower risk patients.
At the beginning of our study, there was no report on the relationship between FN and COVID 19 outbreak, however, several recent research results have been reported. The result of each study though similar in some aspect, differed as highlighted in table 5 (19-23). Nessle et al. reported only a reduced incidence of upper respiratory infection associated FN, and not overall FN episodes among pediatric patients with cancer during post-COVID 19 era (19). Three other retrospective studies showed a reduction in the overall FN incidence after COVID 19 outbreak in relatively larger dataset. However, these data were derived from heterogeneous disease groups with different tumor types, chemotherapy regimen, patient’s characteristics (20, 21, 23). Furthermore, Baracy et al. reported an unexpected finding that FN episodes were reduced only in patients with hematologic malignancies (21). Gwak et al. reported a reduction in FN episode in a homogenous patient group that received adjuvant docetaxel, adriamycin and cyclophosphamide chemotherapy after breast cancer surgery (22). This study was very similar to ours, but enrolled a relatively smaller number of patients, and conducted somewhat different treatment and preventive strategies from that of routine clinical practice. Since all the studies were retrospective design, missing data bias is important limitation to interpreting their results. Since the hospital where our study was conducted is the only university hospital on Jeju island, and almost all patients are referred to our center, missing data bias is expected to be minimal.
Our study has some limitation. A retrospective, observational study with small number of patients may have some potential bias. Another limitation is that the change in clinical practices after COVID 19 outbreak (e.g., fewer hospital visits due to patient’s hesitation or higher institutional barrier to hospital visit) could have affected the reported FN episodes. However, a careful review of medical records would have considerably solved this problem. As our study enrolled a very limited patient population, these results cannot be applied to general patients with cancer with diverse clinical characteristics. We believe that the result of this study can be applied to patients with lower FN risk.