Understanding oncologic emergencies and related emergency department visits and hospitalizations: a systematic review

Study design

We conducted a systematic review of the literature, which was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [28].

Eligibility criteria

We included original research articles that examined oncological emergencies and interventions. We included all cancer types and did not require the publications to articulate strict diagnostic criteria or treatment interventions. We excluded non-English-language publications.

Search strategy

Figure 1 illustrates the detailed search strategy. We searched PubMed, Web of Science, and EMBASE healthcare databases for literature published between January 1st, 2003, to December 31st, 2022. A liaison librarian (JM), in collaboration with the research team, developed the publication search strategy to identify publication abstracts corresponding to the following Medical Subject Headings (MeSH) and keywords: oncologic emergencies, emergency medicine, acute care, and cancer. This search strategy was adapted for each database (e.g., EMBASE). A second liaison librarian (DC) performed a peer review of electronic search strategies (PRESS) to refine the search further. Investigators applied the refined search strategy to additional databases which identified 12,491 citations. Investigators imported these results in Covidence (https://www.covidence.org), a web-based software platform for conducting systematic reviews that additionally eliminates duplicate records

Study selection

Studies were included if they were (1) empirical, peer-reviewed original research, (2) focused on adult (≥ 18 years) patients with cancer, (3) on cancer as a primary disease in the ED, ICU, hospital, other urgent or acute care settings, (4) included chief complaints at ED or predictors of ED use and had explicitly defined oncologic emergencies [4]. Investigators excluded: (1) abstracts, literature reviews, editorial reviews/commentaries, conceptual papers, or case reports; (2) studies that explored pediatric populations (as part of or the entire sample), and (3) referenced oncologic emergencies without subsequent clinical address or care focus. A group of 12 reviewers (SY, AH, KA, JK, NW, MH, JB, MW, BGR, SH, CRG, CC) screened records from the initial search of the databases and agreed on studies for inclusion. Each record underwent an initial title/abstract screening and a subsequent full-text screening, with each reference reviewed by at least two coders. The research team resolved discrepancies or ambiguities in coding during consensus meetings.

Data extraction and analysis

A study-specific abstraction coding sheet was developed a priori. The study design, country where the study was conducted, number of sites, whether it was an ED or any other acute care setting (e.g., ICU), number of patients, cause of the ED visit/hospitalization (i.e., oncologic emergency specified), intervention(s) used to treat that oncologic emergency, and the outcomes studied were abstracted. Other extracted data included aim and objectives, methods, study population, nature of oncologic emergencies and intervention used in the ED/hospital, outcome measures, results, and conclusions. Given our focus on various types of empirical studies and the exploratory nature of this review, quality assessment was not performed. However, studies were included only if the publication reported specific oncologic emergencies and/or interventions used in the ED. Figure 2 illustrates the review process.

Description of studies

Table 1 notes study characteristics and Table 2 describes the articles in detail [29–77]. The search strategy identified 12,491 citations about cancer-related emergency visits, of which 49 met inclusion criteria (Fig. 2): 28 retrospective cohort studies, 13 prospective cohort studies, four randomized controlled trials (RCTs), three cross-sectional, and one case–control design. Of the papers reviewed, 18 originated from the Americas, 13 were from Europe, 12 were from Asian countries, two from Australia, and four from multi-national studies (Fig. 3). Most studies were conducted in a single site (n = 34/49), with comparable proportions of ED visits (n = 22/49) and acute care hospital/ICU visits (n = 27/49). The study population size varied widely among the identified studies (range 9 – 87,555 participants), with most studies not specifying the cancer type, stage, or treatment types, and over 63% of studies had an average age ≥ 60 years. The earliest publication meeting the study criteria appeared in 2003, with 84% of the remaining papers published after 2010.

Oncologic emergencies

The most common oncologic emergencies or predictors of ED use/hospitalization included infection (n = 22/49), pain (n = 20/49), dyspnea or respiratory symptoms (n = 19/49), gastrointestinal (GI) symptoms, including bowel obstructions (n = 17/49), and treatment-related toxicities (n = 13/49). Interventions within the ED or hospital ranged from pharmacological management with opioids (n = 11/49), antibiotics (n = 9/49), corticosteroids (n = 5/49), and invasive procedures (e.g., palliative stenting; n = 13/49) or surgical interventions (n = 2/49).

Management of infection

Infections included sepsis, febrile neutropenia, and pneumonia. Isolated cytokine release syndrome and sepsis caused by pneumonia, enterocolitis, and skin infections in patients with hematological malignancies were managed with life-saving treatments such as vasoactive drugs, non-invasive ventilation, and corticosteroid therapy [31]. Bou Chebl et al. [34] also addressed sepsis, utilizing interventions including IV fluids, vasopressors, and intubation. Some studies which focused on febrile neutropenia in patients with hematological malignancies [35, 56]did not specify ED interventions, while other studies [42, 65] highlighted antibiotic therapy as a crucial treatment. In patients with bacterial pneumonia, Gudiol et al. (2016) [46] noted the necessity for ICU admission, mechanical ventilation, and targeted antibiotic therapy. Additionally, Grewal et al. [45] emphasized the need for specialty consultations and hospital admissions for infections in cancer patients.

Addressing pain

Pain is a common and debilitating symptom in oncological emergencies, often managed with various analgesic strategies. Mercadante et al. (2010, 2017) [53, 54] reported on the management of pain using opioids such as intravenous morphine, oral morphine, transdermal fentanyl, and oral oxycodone. When comparing the efficacy of intranasal fentanyl (INF) versus intravenous hydromorphone (IVH) for severe pain relief, they report that IVH provided superior pain management [32]. Coyne et al. (2021) [40] and Patel et al. (2017) [58] both discussed opioid therapy for pain management, with Coyne et al. (2021) [40] also mentioning acetaminophen as an adjunct treatment. The use of various medications, including opioids and corticosteroids, provided acute symptom management [52]. When explicitly addressing the palliation of malignant large-bowel obstruction in colorectal cancer, palliative stenting or stoma creation was used by Abelson et al. [29]

Assessment of dyspnea and respiratory symptoms

Dyspnea was a focus of several studies. Chou et al. (2012) [38] examined sepsis-related respiratory failure in stage III and IV lung cancer patients, managing the condition with mechanical ventilation and transferring patients to long-term respiratory care if necessary. Cooksley et al. (2020) [39] addressed dyspnea related to immune checkpoint inhibitor toxicity, using steroids and antibiotics as interventions. In patients with advanced lung cancer, Kim et al. (2014) [50] identified multiple causes of dyspnea, such as obstructive pneumonia and chemo-induced lung toxicity, managed through vasoactive agents, hemodialysis, and mechanical ventilation. Additionally, Yilmaz et al. [77] explored the management of dyspnea in lung cancer patients with salbutamol and magnesium sulfate. Peyrony et al. (2021) [59] reported various supportive measures, including oxygen therapy and mechanical ventilation.

Addressing GI symptoms & bowel obstruction

GI symptoms included bowel obstruction, diarrhea, colitis, and nausea. Abelson et al. [29] and Frago et al. (2010) [43] both studied malignant large-bowel obstruction in colorectal cancer, which was treated with palliative stenting or chemotherapy regimens. Castillo et al. (2021) [36] focused on immune-related adverse events such as colitis and diarrhea, which were managed with corticosteroids including dexamethasone and prednisone. Teimouri et al. [68] also highlighted the use of steroids for managing GI toxicities, including colitis and hepatitis. In patients experiencing GI symptoms in the ED, Delgado-Guay et al. (2015) [41] noted the importance of palliative care consultations. Waddle et al. [71] reported on managing various GI symptoms and associated complications without specifying particular interventions.

Other treatment-related toxicities

Treatment-related toxicities are critical oncological emergencies, often requiring immediate intervention due to their potential severity and impact on patients' quality of life. Castillo et al. (2021) [36] explored immune-related adverse events (irAEs) such as colitis, diarrhea, hyperglycemia, and shortness of breath, managed with irAEs-specific treatments, primarily corticosteroids. Similarly, Teimouri et al. [68] focused on managing toxicities like diarrhea/colitis, hepatitis, pneumonitis, nephrotoxicity, and cardiotoxicity with steroids. In patients experiencing dyspnea, diarrhea, and fever related to treatment toxicity, Cooksley et al. (2020) [39] employed corticosteroids and antibiotics as treatment modalities.

Kim et al. (2014) [50] demonstrated a range of toxicities in lung cancer patients, including chemo-induced lung toxicity and radiation pneumonitis, addressing these conditions with interventions including vasoactive agents, hemodialysis, and mechanical ventilation. They emphasized that this comprehensive approach is vital to managing the acute and potentially life-threatening effects of cancer treatments. Xia and Wang (2016) [74] also reported on severe treatment-related toxicities in solid tumor patients, such as respiratory failure, septic shock, and acute renal failure, which necessitated the use of vasopressors, mechanical ventilation, and renal replacement therapy.

Lagman et al. [52] discussed the acute inpatient management of treatment-related complications, including neutropenic fever and radiation-induced pneumonitis, highlighting the use of various medications, invasive procedures, and supportive care measures to stabilize patients. Additionally, Abelson et al. [29] described interventions for malignant large-bowel obstruction secondary to colorectal cancer treatments, including palliative stenting or stoma creation.

Strengths and limitations

Our study identified diverse interventions used in the ED and other acute care settings for cancer-related concerns. We demonstrated a paramount need for palliative, hospice, and advanced care planning among this population. We have noted the need to strengthen endeavors that promote the integration of palliative care into standard emergency care for patients, especially those with cancer.

We restricted article inclusion to English language reviews and those reviews published in academic journals. Regarding the articles included, most studies did not report cancer type and treatment type, and we are unaware of the time required for intervention from these studies. Thus, we are unable to extrapolate the probability of emergent care according to the cancer type and complication. Most studies were conducted in a single site, which may limit the generalizability of findings due to variations in local healthcare practices, resources, and patient demographics. The heterogeneity in patient populations, with a significant proportion of studies not specifying cancer type, stage, or treatment, suggests a need for more granular data better to understand the specific needs of different patient subgroups.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

Bonnie E. Gould Rothberg: Stock Ownership: Butterfly Networks Inc., Quantum Si, Hyperfine Research, AI Therapeutics, Detect Labs, identifeye Health, Protein Evolution Inc.) and direct family members: #A) Stock Ownership: Butterfly Networks Inc., Quantum Si, Hyperfine Research, AI Therapeutics, Detect Labs, identifeye Health, Protein Evolution Inc., 454 Corporation, Electric Futures, (Abbvie Inc, Amgen inc, Biocryst Pharmaceuticals, Gilead Sciences, Inc, Regeneron Pharmaceuticals Inc, Roche Holdings, Pacific Biosciences of California Inc (sold 07/2023)), Telesis Bio Inc #B) Employment: Butterfly Networks Inc, 4Catalyzer #C) Significant Leadership: Butterfly Networks Inc, Quantum Si, Hyperfine Research, AI Therapeutics, identifyee Health, Detect Labs, Protein Evolution Inc. #D) Patents, Royalties, and Other Intellectual Property: Thermo Fisher, Butterfly Networks, Quantum Si, Hyperfine Research, AI Therapeutics, Tesseract, Detect Labs. Christopher William Baugh: paid speaker for Roche Diagnostics, Octapharma, and CE Symmetry, an investigator for Abbott Laboratories, an advisory board participant for Roche Diagnostics, Salix Pharmaceuticals, Pfizer Inc., and AstraZeneca, a consultant for Abbott, Pfizer, Roche, and an advisor to Lucia Health Guidelines. Sai-Ching Yeung: an advisory board participant Salix Pharmaceuticals. The remaining authors declare that they have no competing interests.

Christopher William Baugh: paid speaker for Roche Diagnostics, Octapharma, and CE Symmetry, an investigator for Abbott Laboratories, an advisory board participant for Roche Diagnostics, Salix Pharmaceuticals, Pfizer Inc., and AstraZeneca, a consultant for Abbott, Pfizer, Roche, and an advisor to Lucia Health Guidelines.

Sai-Ching Yeung: an advisory board participant Salix Pharmaceuticals.

The remaining authors declare that they have no competing interests.