Hematological cancer or malignancies are cancers of the blood, bone marrow, and lymph nodes that come from either lymphoid cell lines or myeloid. Megakaryocytes, macrophages, granulocytes, mast cells, and erythrocytes are produced by the myeloid cell line. The myeloid cell line is responsible for acute and myelodysplastic syndromes, myeloproliferative disorders, and chronic myelogenous leukemia [1]. Other cells such as plasma, T and B cells, as well as natural killer cells or large granular lymphocytes are produced by the lymphoid cell line. This lymphoid cell line is responsible for lymphomas, lymphocytic leukemia, and myelomas [1]. In the United States and United Kingdom, 16 (13.6%) leukemia cases are reported per 100,000 people [2,3]. The clinical course and prognosis vary as it may depend on the existence or the type of genetic mutation itself. Chemotherapy, radiation, immunotherapy, and bone marrow transplantation are examples of active treatment options.

A healthy individual with a strong immune system is usually resistant to opportunistic infections (OIs) caused by pathogens such as protozoa, fungi, viruses, or bacteria [1]. However, if the immune system is weakened, the pathogens have a better chance of infecting the individual. Cancer therapy to treat hematological cancer can weaken a patient’s immune system, making them vulnerable to OIs. OIs affect 5%-60% of individuals with hematological malignancies [4-7], with the mortality rate ranging between 15% and 41% [5]. Broad-spectrum intravenous antimicrobials are used to treat OIs, which are first used empirically and then adapted to the patient’s specific needs [8]. Viruses causing the OIs can be treated with antiviral therapies.

Most individuals with hematological cancers are more susceptible to OIs due to both immunological impairment of cancer therapies and disturbances caused by the disease itself [9,10]. In a situation where patients with cancer become neutropenic, they are isolated as a precautionary measure to prevent OIs, while the length of stay in the isolation room varies depending on their medical condition. This may cause wide range of psychological burdens such as depression, anxiety, and stress as a result of staying in the isolation room [11]. It has been reported that the psychological well-being of patients with cancer influences their treatment response and long-term prognosis [12-14]. Nondrug therapies such as isolation have been used to reduce the OIs. The relation between OIs and cancer treatment is bidirectional. Treatment for various types of cancers causes immunosuppression and that may cause OIs among patients with cancer. Treatment knowledge, understanding, and adherence will improve the overall prognosis [15,16]. Therefore, the objective of this review is to evaluate the efficiency of nondrug interventions in preventing OIs in individuals with hematological cancers. As a result, this study was performed to assess the safety and efficacy of nondrug interventions for the prevention of OIs in individuals with hematological cancer or malignancies.

There are three major types of nondrug interventions. The first type of intervention is the barrier method to prevent potential transmission such as wearing protective equipment or gloves, cleaning of bed sheets and clothes, and the use of masks. The second type of intervention is complete elimination of the causal agents such as fumigation on a regular basis, regular cleaning of potential microbe-harboring goods including toys and carpets, elimination of houseplants that are likely to be a reservoir for microorganisms, and the use of mouthwash as a personal hygiene modification [17]. The third type of intervention is the physical method such as applying positive pressure (controls the air quality in-flow) or using a high-efficiency particulate absorption (HEPA) filtered room to improve the hospital environment.

In patients with neutropenia, primary infections can be a result of mild injuries caused by venous and vascular catheters, which can spread through the bloodstream and eventually result in soft tissue and skin infection [18]. Neutropenic diet is also a part of the barrier method to reduce the risk of infection. It consists of a low-bacteria diet, such as meals cooked thoroughly or with boiled water. Despite its widespread use, the effectiveness of a neutropenic diet in patients undergoing chemotherapy remains debatable [19-21].

Physical methods such as using an HEPA filter may help prevent contact with pathogens that exist in soil or plants and can reduce OIs in vulnerable individuals [22-26].

Personal hygiene modifications, such as the use of mouthwash [27], chlorhexidine baths [28], and frequent cleaning of surfaces, may decrease the microorganisms and completely eliminate them [29].

Targeted nondrug interventions to treat OIs act in 3 ways: removing the cause of illness, reducing the contact with infectious agents, and decreasing the risk of microorganism invasion.

The use of personal protective equipment (PPE) and HEPA filters are examples of barrier methods to reduce one’s exposure to an infective agent. HEPA filters are reported to eliminate 99.97% of particles with a diameter over 0.3 m, which includes most microorganisms [30]. The use of PPE such as surgical and N95 masks has demonstrated beneficial effect in reducing the transmission of influenza virus [31,32]. Hand hygiene has been shown to prevent the transfer of harmful microorganisms, such as central line–acquired bloodstream infections and methicillin-resistant Staphylococcus aureus [33,34]. The use of gloves may limit the quantity of microbes’ transmission through skin contact. Alcohol-based antiseptics have been suggested to reduce pathogen transmission by denaturing the microorganisms’ proteins [1].

OIs contribute to mortality and morbidity in individuals with hematological cancer. In these patients, nondrug interventions, such as the use of PPE and modifying the environment, are commonly used, as they may lessen the need for medication prophylaxis and therapy; however, they are not without hazards and expenses. As a result, synthesizing the existing evidence on the safety and efficacy of these interventions is critical. In that regard, this study was performed to assess the safety and efficacy of nondrug interventions for OI prevention in individuals with hematological cancer or malignancies.

We conducted a systematic review by following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [35]. The PRISMA checklist was used while writing this report [36].

For this review, we followed the PICOS mnemonic, where “P” represents adult patients with hematological cancer, and we included all individuals or patients regardless of the stage of disease, type of hematological cancer or malignancy, morbidity, and the received treatment modality; “I” is any nondrug intervention included either alone or in combination with other therapies such as alteration of patient/caregiver behavior, alteration of the home-based environment, and hospital-based environmental control measures. We also included other interventions defined by the study authors as nondrug intervention. The control “C” group is defined as individuals or patients who did not receive the nondrug intervention, or those who obtained prophylactic pharmacological medications or therapies. The outcomes “O” are defined as either OIs or bacterial OIs, mortality due to OIs, all-cause mortality, hospitalization duration in days, chemotherapy interruption (number of episodes or duration of interruption), quality of life, and adverse effects related to the intervention. The included study design “S” is all randomized controlled trials (RCTs) or cluster RCTs that are published in full texts or abstracts. We excluded studies when the intervention contained pharmacological measures or any studies with the crossover design due to concerns of the “carryover” effects.

A comprehensive systematic literature search was performed on various electronic databases, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), WHO (World Health Organization) International Clinical Trials Registry Platform (ICTRP), and PubMed, to identify relevant studies from inception to December 2022. A strategy search for PubMed was established and modified for use in the other databases. During the search, keywords and equivalent MeSH (Medical Subject Headings) phrases were combined when applicable, with no language or publication year restrictions. The search strategies for MEDLINE are presented in Multimedia Appendix 1 and CENTRAL in Multimedia Appendix 2. We also searched the abstract records from the following conferences organized by societies that are related to blood cancer or malignancies: The European Society for Medical Oncology Annual Congress, The European Hematology Association (EHA) Conference, American Association for Cancer Research and American Society of Clinical Oncology Conference, American Society of Hematology Meeting, and Virginia Association of Hematologists and Oncologists (VAHO) Spring Membership Conference. We also searched the following databases for ongoing studies: metaRegister of Controlled Trials [37], International Clinical Trials Registry Platform [38], and ClinicalTrials.gov [39]. During the searches, there was no limitation or restriction on the language of the article.

A total of 4700 records (1746 records from CENTRAL [Cochrane Central Register of Controlled Trials], 1067 records from MEDLINE, and 1887 records from Embase) were successfully retrieved. No additional records were identified through other sources, such as online conference archives and clinical trial registries. After removing duplicates, 4641 records remained. Subsequently, 4615 records were excluded. We obtained 26 records to be assessed for eligibility. Of these, 20 articles were excluded. Ultimately, 6 eligible articles or studies were included. The PRISMA flow diagram of the selection of studies is depicted in Figure 1.

The 6 included studies were published as full papers [42-47]. Of these 6 studies, 2 were performed in Germany [42,45], 1 was conducted in Canada [43], 1 in Finland [44], 1 in The Netherlands [46], and 1 in Iran [47]. Table 1 describes in detail the characteristics of the studies included in this review.

We excluded 20 out of the short-listed 26 studies [48-67] as depicted in Multimedia Appendix 5. The studies were excluded due to the following reasons:

We did not identify any on-going study for this review.

Overall, the risk-of-bias profile of the included studies varied, with insufficient information in most studies to enable a meaningful assessment of the risk-of-selection bias, and the high risk of performance bias in half of the included studies was due to a lack of blinding of individuals and personnel. The distribution of risk of bias in different aspects for the included studies are shown in Figures 2 and 3 and Multimedia Appendix 3.

A total of 519 participants were evaluated from 6 included studies. Although this review included a small number of studies, it represented the best existing evidence that addressed the use of nondrug intervention for OIs. This review identified 4b major types of nondrug interventions for hematological cancer: (1) mouthwash that contained either chlorhexidine, nystatin, saline, amine fluoride-stannous fluoride, sodium fluoride, or herbal substances; (2) CVCs that were coated with chlorhexidine silver-sulfadiazine or uncoated; (3) use of well-fitted masks; and (4) diet consumed (either a low-bacterial diet or a normal diet). However, each type of intervention was represented by 1 small study. Overall, the use of chlorhexidine mouthwash alone or in combination with nystatin or nystatin mouthwash alone showed no clear difference for reductions in mucositis based on mucositis score and ulcer size [43]. A study by Laine et al [44] assessed the use of amine fluoride-stannous fluoride or sodium fluoride mouthwash. The study, however, did not show any clear difference in all-cause mortality and adverse effects such as discomfort in the mouth, teeth staining, unpleasant taste, and nausea. Maschmeyer et al [45] assessed the use of a well-fitted mask with no clear difference in OIs for all-cause mortality. Another study by Tiel et al [46] evaluated the use of a low bacteria diet versus a standard diet with no clear difference in the reduction of possible OIs or OIs assessed clinically. A study by Ardakani et al [47] that assessed the use of herbal mouthwash in preventing OIs showed no clear difference when compared with the placebo mouthwash group.

We found 2 published reviews that examined the effectiveness of nondrug interventions in preventing OIs among different populations. Helder et al [68] had assessed the effectiveness of 5 different nondrug interventions in preventing bloodstream infections among newborns admitted to a neonatal intensive care unit. The authors included 15 RCTs and found that proper CVC insertion and maintenance with a proper aseptic technique were the most effective interventions to prevent bloodstream infection in infants. The review differed from our review, as ours focused on evaluating the outcomes among adults with hematological malignancies. The other review by Wekesah et al [69] assessed the effectiveness of nondrug interventions in improving outcomes and quality of care among pregnant women in sub-Saharan Africa. The authors included 73 mixed design studies and identified many interventions for improving maternal health. The review differed in scope from our current review, as we only focused on the prevention of OIs among individuals with hematological cancers.

To the best of our knowledge, there is no systematic review that evaluated the use of nondrug interventions in preventing OIs among individuals with hematological cancer. The only available reviews that assessed interventions relevant to our review are those that evaluated antimicrobial-coated CVCs. One Cochrane systematic review and a related meta-analysis assessed the safety and effectiveness of antimicrobial catheters for patients in the intensive care unit, hematological and oncology unit (with all types of malignancies), and community settings [44,45], and reported that antimicrobial catheters in general reduced catheter colonization without clearly reducing catheter-related bloodstream infections, overall sepsis, and mortality rates. This is consistent with the finding of our single included study [42], which was also included in both reviews.

The review also has strengths. First, this is the first systematic review that evaluated the use of nondrug interventions for preventing OIs among patients with hematological cancers. We confirmed the effectiveness through the synthesis of evidence, and the result has significant clinical implication for both oncologists and patients. Second, we established strict inclusion and exclusion criteria, which resulted in a uniform data to be evaluated. We also included our assessment on study quality, which allows readers to judge the strength of evidence.

The review has a few limitations. First, despite our comprehensive search, we could find only very few studies related to OIs that are eligible to be included. This review also only focused on the effect of nondrug interventions for hematological cancers that limited the applicability of its results to other types of cancer or immunocompromised patients. Further, we might have missed relevant papers from smaller databases, especially those that are non-English. In addition, there might be publication bias that we were unable to rigorously evaluate as the number of included studies was too small. Besides, we only included RCTs in our review with a well-defined, relatively narrow set of patient population, and as such, serious or rare adverse events might not have been comprehensively captured.

Overall, the quality of evidence presented in this review was very low. We are uncertain on the efficacy and safety of various types of mouthwash, coated CVCs, use of well-fitted masks, and low bacterial diet in major clinical findings such as OIs and related outcomes. Insufficient evidence exists on the effect of the nondrug intervention for preventing OIs in people with hematological cancers, and in people who are immunocompromised. This lack of evidence should be kept in mind when balancing the beneficial effects of nondrug interventions against the cost and feasibility of implementation in specific settings and against the potential for the development of OIs, and thus no firm conclusion can be made to inform practice. Therefore, further evidence is needed regarding the effect of nondrug interventions in patients with cancers, or in those who are immunocompromised.