Letter to the Editor: Toward better mortality prediction in intensive care unit-admitted colorectal cancer patients

Abstract

Colorectal cancer (CRC) remains a leading cause of cancer death worldwide. Dong et al evaluated prognostic factors associated with short-term outcomes among intensive care unit (ICU)-admitted CRC patients and highlighted the lack of a reliable prognostic tool despite commonly used ICU scores such as Sequential Organ Failure Assessment and Acute Physiology and Chronic Health Evaluation II. Their multicenter cohort (n = 189) reported a 90-day ICU mortality of 12.2%; however, modest sample size limits generalizability. The study by Dong et al, published in the recent issue of the World Journal of Gastrointestinal Oncology, links higher Sequential Organ Failure Assessment scores and emergency surgery to increased 90-day mortality in stage III disease. We argue that a robust prognostic model for this population should be multidimensional – incorporating frailty, cancer stage and treatment context (elective vs emergency surgery), comorbidities, nutritional and socioeconomic status, and timing/acuity of ICU transfer. Prospective, adequately powered studies are needed to develop and validate such tools to guide ICU triage and optimize outcomes for CRC patients. This manuscript call attention to the urgent need for validated, multidimensional prognostic models to guide ICU decision-making in CRC care.

Key Words: Colorectal cancer; Intensive care unit; Prognostic factors; Frailty scores; Stage of disease

Core Tip: Colorectal cancer remains a leading cause of cancer-related mortality globally. While screening and treatment have increased disease prevalence, reliable predictors of 90-day mortality in intensive care unit-colorectal cancer treated patients are lacking. Prognostic scores such as Acute Physiology and Chronic Health Evaluation II and Sequential Organ Failure Assessment show potential but require further validation. Additionally, factors like frailty, cancer stage, socioeconomic status, and surgical urgency may influence outcomes/prognosis. Currently, no single tool adequately captures the complexity of prognosis in this population, highlighting the need for prospective studies to develop a more comprehensive and accurate prognostic model.

TO THE EDITOR

Colorectal cancer (CRC) remains one of the leading causes of morbidity and mortality worldwide, representing a significant global health burden. Over the past years, substantial progress has been made in cancer care through early screening programs, multimodal treatment strategies, and follow-up protocols. These efforts have contributed to improved outcomes and survival rates for many patients. However, a critical and often underexplored aspect of CRC care involves the management of patients who require intensive care unit (ICU) admission. Despite the growing population of CRC patients undergoing active treatment and the need for ICU care, no clear prognostic score currently exists.

Patients with malignancies, including CRC, are generally more vulnerable due to underlying immunosuppression, frequent hospitalizations, and the systemic effects of both the disease and its treatment. These factors place them at increased risk for acute clinical deterioration, often necessitating escalation to ICU-level care. Infections, sepsis, postoperative complications, tumor-related emergencies (e.g., bowel perforation, obstruction, or hemorrhage), and chemotherapy-related toxicities are just some of the common triggers for ICU admission in this population. Despite the prevalence of such events, there is currently no standardized scoring system or reliable set of predictive factors to guide clinicians in anticipating ICU admissions or forecasting mortality risks specifically among CRC patients. This lack of prognostic tools hinders timely decision-making and may contribute to suboptimal outcomes. Developing validated models or criteria could help identify high-risk patients earlier, ultimately improving survival and quality of life[1,2].

Given the rising incidence of CRC and the increasing complexity of care, there is a clear and urgent need for more research focused on ICU utilization, outcomes, and predictive tools tailored to this population. Addressing this gap could play a pivotal role in advancing comprehensive, patient-centered care in oncology. It is clear that there is a need for both an intensivist and a multidisciplinary team capable of caring for this type of patient[3,4]. It is also important to note that postoperative care should not be the sole reason for admission to ICU for CRC patients, as many complications or the need for the ICU level of care for this type of patient can arise early in their hospital course or be due to another reason, like the effects of the treatment[1,4]. Overall, the number of patients undergoing active treatment and survivors of CRC is going to increase in the upcoming years, thus the need for an adequate prognostic tool and markers for both ICU admission and 90-day ICU mortality. The creation of a score can facilitate an early and adequate upgrade and then care in the ICU if needed.

DISCUSSION

A prognostic tool for 90-day ICU mortality in CRC should integrate key clinical and patient-specific factors. The care of cancer patients in the ICU differs from that of non-cancer patients, as physicians must manage both the critical illness and the underlying malignancy simultaneously. This highlights the importance of a multidisciplinary approach and close co-management between oncologists and intensivists. A previous retrospective cohort study conducted from Brazilians ICUs found that co-management – along with the involvement of a clinical pharmacist and the implementation of treatment protocols – significantly improved survival rates among cancer patients[4]. While several prognostic tools are available and can be useful, they remain limited, as they often fail to account for all factors involved in the care of cancer patients, particularly those specific to CRC. This discussion further explores the factors that appear to influence outcomes and emphasizes the need to incorporate them into future research.

Historically, scoring systems like Sequential Organ Failure Assessment (SOFA) and Acute Physiology and Chronic Health Evaluation II have served as prognostic indicators of mortality and morbidity and have been widely used in ICU settings. However, the short-term prognosis of CRC patients in the ICU still needs to be investigated[2,5-7]. Recently, the study by Dong et al[2], published in the recent issue of the World Journal of Gastrointestinal Oncology, stated that the absence of a reliable prognostic tool, but well-known scoring systems, such as SOFA and Acute Physiology and Chronic Health Evaluation II, may have some roles in predicting short-term mortality, and the need for further prospective longitudinal studies. Their data were collected among 37 cancer centers, but only 189 patients were included after the appropriate screening, which can significantly decrease the study's power and limit its generalizability. One of the main findings was that the 90-day mortality of CRC patients in the ICU was 12.2%. Also, it appeared that patients admitted to the ICU from the floors/wards may have a higher risk of mortality. Due to the complexity of CRC, an appropriate scoring system must include, but not be limited to, the following factors: Frailty, socioeconomic status, stage of the cancer, treatment modality received, other comorbidities, and the need for urgent surgery, among other relevant markers.

Prior research has shown that frailty in the hospital setting is linked to a higher risk of death, greater use of healthcare services, and increased disability. Frailty also has the potential to help identify high-risk patients more effectively in the ICU. Historically, the Clinical Frailty Scale, a metric that incorporates a patient’s physical activity, functional status, chronic illnesses, and cognition, has served as a valuable tool to predict prognosis in the ICU. Patients with higher scores have higher numbers of complications, disease burden, and usually tend to have poorer outcomes in the ICU[7,8]. Also, a study by Welford et al[8] found that the clinical frailty score could help predict the overall prognosis for inpatient cancer patients. Frailty is linked to a higher risk of complications during oncological surgery and chemotherapy, which can lead to a worse patient's outcome[8]. Similarly, a meta-analysis by Cai et al[9] concluded that the efficacy of frailty screening in CRC patients lies in its ability to predict overall prognosis, which can help guide early treatment for this group of patients. Overall, we believe that frailty can be incorporated into a scoring system for predicting the prognosis and mortality of CRC patients in the ICU setting.

Another important tool to include in a possible predictive score or factors influencing CRC patients’ ICU mortality should include the stage of disease and treatment modality the patient is undergoing. Patients with more advanced stages of the disease, those receiving palliative chemotherapy or emergent local surgical resection due to obstruction or perforations, are already sicker patients at baseline. In a recent study, Kim et al[6] reported that stage III CRC patients who underwent emergency surgery and were in the ICU had higher SOFA scores and mortality within 90 days[8]. This relationship was deemed statistically significant. Colonic perforations due to tumor size or growth most of the time lead to emergency surgery, which has also been shown to have an association with worse overall outcomes. In these cases, an adequate oncological surgical resection may not be possible, leading to increased complications down the line[10]. All of these can be independent factors that can serve as predictors of mortality of CRC in the ICU setting.

Socioeconomic status, nutritional state, comorbidity burden, family support, and institutional surgical volume also affect outcomes and should be included in model development[11]. Additionally, it is noteworthy that planned vs unplanned ICU transfers have been shown to have different survival outcomes[2]. It remains unclear which factors play the most significant role in determining 90-day survival for CRC patients admitted to the ICU: (1) Those related to CRC; (2) Those associated with the acute illness; and (3) Patient-specific factors. Thus, it is necessary to consider all the above factors when creating a 90-day mortality prognostic indicator.

CONCLUSION

Overall, each CRC patient has their own specific conditions that can affect overall outcomes in the ICU. As the population of CRC patients undergoing active treatment increases, there is a need for an adequate prognostic tool, marker, or score that can help predict early mortality after an ICU admission. This score should include variations of the following: Frailty, stage of disease, emergent vs elective surgery, and socioeconomic factors. To develop a better prognostic tool that captures each patient's individual factors we need prospective longitudinal studies that examine all these variables. It is crucial to develop and validate a multifactorial prognostic tool to improve triage, optimize ICU resources, and enhance survival in CRC patients.