Intensive care unit prognostic factors in critically ill patients with advanced solid tumors: a 3-year retrospective study

Life expectancy is rising globally and the incidence of all-types of cancer is predicted to increase from 12.7 million new cases in 2008 to 22.2 million by 2030 [1]. An increasing number of older patients will live with tumors and acquire life-threatening complications from radical surgery, high-dose chemotherapy, adverse drug events [2], increased susceptibility to infection [3, 4] or cancer itself (such as tumor lysis syndrome, and hypercalcemia of malignancy) [5, 6]. As a consequence, there is an increase in critically ill patients with various types of malignancy at any stage requiring intensive care.

Cancer treatment near the end-of-life has become more aggressive and intensive care unit (ICU) mortality of cancer patients has improved in recent years [7‐13]. However, patients with hematological or advanced-stage solid malignancies are still frequently denied admission to ICUs according to current policies, even if some of them may survive. Selection of patients inevitably leads to undertreatment and unnecessary deaths [11].

There have been few studies about unplanned ICU admission of critically ill patients with advanced solid tumors in China; therefore, we conducted this study to establish independent risk factors for prognosis in this patient subgroup. Three classic scoring systems on different lengths of stay in the ICU were compared for predicting prognosis. It will be helpful to identify patients who are most likely to benefit from critical care and decide the best time to terminate the ICU trial and discuss a change in code status.

Recently, Gruber and co-workers reported a 12-month mortality rate of 48.3 % for long-stay ICU patients with cancer, which means that more than half of long-stay critically ill cancer patients survive ≥1 year [10]. Many studies have documented improved survival of critically ill patients with cancer. Two main hypotheses have been proposed to account for the decreased mortality rate. First, the development of more potent and targeted anti-tumor therapies, advances in the standard strategies for determining indications and supportive care, as well as progress in the prevention of organ dysfunction. Cancer patients benefit from reduced cancer-related complications or timely intervention. Second, with a deeper understanding of the pathophysiological mechanisms in organ dysfunction, intensive care has improved survival of critical illness by constantly renewing strategies for survival of sepsis, hemodynamic monitoring, mechanical ventilation, nutrition support, sedation, and analgesia [5, 12, 16].

The in-hospital mortality of patients with solid cancer in our study was similar to that reported from European ICUs [4]. The crude ICU mortality was 14.9 %. The ICU mortality of patients diagnosed with septic shock was 26.7 and 10 % in those diagnosed with severe sepsis. The ICU mortality of patients who required vasopressors, mechanical ventilation or renal replacement therapy for >24 h was 25, 25.9 and 40 %, respectively. When patients were admitted to the ICU, their APACHEII or SOFA scores were comparable to those from most previous studies. However, the reason why the ICU mortality rate observed in our study mentioned above was lower than previously 30–70 % was multifactorial [4, 17‐21], including different underlying diseases, types of cancer, and ICU admission or discharge criteria. As patients with early-stage solid tumors after elective surgery were the main group in our ICU, the higher mortality of cancer patients admitted for medical reasons was also observed (14.9 % vs 2.1 %) [21].

As intensive care specialists, we should realize that the endpoint of therapy in patients with advanced-stage cancer differs from that in patients without cancer. We should not be concerned only with survival rate but also with long-term survival and quality of life [6]. During our study, in-hospital survival reached nearly 70 % after a median 6 days in the ICU. APACHEII score on day 1 predicted poor OS, but the mean OS had already reached 28.6 months. Thirty patients (21.3 %) had the opportunity to receive anti-cancer treatment after ICU treatment. Active treatment in the ICU could be more important than many anti-cancer therapies if offers the possibility of prolonging survival with good quality of life for >3 months. In fact, we reached this outcome after a median 6 days of ICU treatment.

Patients with advanced-stage cancer are frequently denied admission to ICUs that are normally run by non-oncologists according to current policy. Several studies have failed to show that diagnosis or stage of cancer is an independent predictor of ICU mortality [4, 9, 10, 16, 18, 19, 22‐24], which was confirmed by our logistic regression model. In other words, triage decisions solely based on the type of cancer are thus not justified. Intensivists sometimes need to make quick decisions based on little or inconclusive information. Sometimes, we may find a high hospital survival rate in a small number of patients for whom an agreement to limit care was not achieved [25]. Thiéry and co-workers showed 26 % survival on day 30 in patients who were considered too ill to benefit from ICU admission. Among the patients who were denied ICU admission because they were felt to be too well to benefit from admission, one quarter were subsequently admitted, and mortality was high (61.5 %) in this subgroup [26]. Rapid selection depending on unreliable triage criteria will inevitably lead to undertreatment and unnecessary death in a minority of patients [11]. The balance between reasonable hope of benefit and excessive burdens on the family or community urgently requires an effective oncology critical scoring system and risk factors analysis to broaden ICU admission criteria for patients with cancer. APACHEIIand SOFA are the most commonly used scoring systems in the ICU, while Eastern Cooperative Oncology Group performance status (ECOG-PS) or KPS is often used in oncology departments to evaluate indications for anti-tumor therapy. SOFA score on day 3–6 in the ICU [4, 17‐19, 23, 27] and ECOG-PS [17, 22, 28, 29] are frequently mentioned as significant risk factors for prognosis. In our study, SOFA score on day 7 of ICU treatment was assessed to be the only significant predictor of ICU mortality, which means that poor performance on admission plays a limited role in the ICU decision-making process. In fact, the severity of physiological derangement in the subsequent 6 days from ICU admission has the biggest impact on ICU survival. We should receive more often than refuse selected patients with cancer for ICU admission [16]. An ICU trial should be offered in particular during the first week of ICU stay [19].

The ICU trial is considered as an alternative to ICU refusal for patients with cancer. It consists of unlimited ICU support, including ambulatory chemotherapy, along with mechanical ventilation and renal replacement therapy, for a limited time period [5, 11, 28]. After the defined 6 days, an interdisciplinary meeting consisting of oncologists, intensivists, nurses, psychologists, and palliative care, pain, and ethics specialists should be held. The treatment goals should shift from curative or supportive therapies to end-of-life care if the reevaluation on day 7 shows clinical deterioration with no available therapeutic options [25]. This decision to limit treatment should be based on certainty of the benefits of the applied treatment and that it does no harm according to the 5th International Consensus Conference in Critical Care [30]. By strengthening the interdisciplinary collaboration to enhance advantages and minimize disadvantages, we could integrate hospice and palliative care with intensive care more effectively and efficiently. That will be the future of oncological ICUs [31‐33].

Our study had several limitations. First, this was a retrospective study at a single cancer center. However, to the best knowledge, it is the first report about the prognosis and risk factors of critically ill patients with advanced solid tumor in the ICU in China. Second, the small size of the sample prevented us from investigating the characteristics of critical illness in patients with different types of solid cancer and the effect of ambulatory chemotherapy. Third, early identification and treatment of critically ill cancer patients on general wards showed no significance in our study, which was contrary to the results from many previous studies. This is probably because the medical emergency team, which facilitates early intervention in response to physiological instability, was not standard in our hospital [14, 16, 18, 23]. The intervention time in our records may have been shorter than the real intervention time. Fourth, ethical consultation is not yet ideal in our hospital. Twenty-one patients in full code status died from an emergency at the end of life because of the lack of ethical consultation at the time of ICU admission. Our results needs to be confirmed by a large prospective study.

In summary, an increasing number of cancer patients require intensive care. The success of active ICU treatment may offer them the opportunity to prolong survival with good quality of life and receive effective anti-cancer therapy. According to traditional ICU admission criteria, critically ill patients with advanced solid tumors are often deprived of the opportunity for intensive care, even though >70 % of them would benefit if admitted. We suggest broadening the criteria for ICU admission. Patients should be allowed an ICU trial that consists of unlimited ICU support, including invasive hemodynamic monitoring, mechanical ventilation, and renal replacement therapy. An interdisciplinary meeting including ethics consultations should be held to make clinical decisions if the SOFA score on day 7 shows clinical deterioration with no available therapeutic options. The goal of the treatment may shift from curative or supportive therapy to end-of-life care.