The Intensive Care Trial for Critically Ill Onco-Haematologic Patients: The Need for Response Criteria at 5 Days of Full Treatment to Separate Good Risk Patients and Avoid Futile Intensive Care Interventions

Introduction: Critically ill onco-hematology patients (pts) admitted to intensive care units (ICU) have poor prognosis. Mechanical ventilation, multiple organ failures and severe sepsis are factors associated with high mortality. Current literature identifies day 5 in ICU as a specific time point at which ceilings of care should be re-addressed.

Patients and methods: We retrospectively reviewed all consecutive onco-haematology patients admitted to the ICU between October 2010 and December 2015. We classified pts according to the reason for ICU admission in 5 groups: a) respiratory failure without mechanical ventilation during the first 24h; b) respiratory failure and mechanical ventilation in the first 24h; c) sepsis without respiratory failure and without renal replacement therapy in the first 24h; d) renal replacement therapy without respiratory failure regardless of septic status; and e) needing hemodynamic support without respiratory failure, sepsis or renal replacement therapy in the first 24h.

After 5 days of full intensive therapy we defined a successful 5-day ICU trial for each of the five groups as follows: a) no mechanical ventilation during 5 days; b) neutrophils > 1.0 or ² 2 organ failures by day 5; c) C-reactive protein decreased by 50% or normalised lactate by day 5; d) off renal replacement therapy by day 5; and e) no inotropic support on day 5. Patients who died during the first 5 days of ICU admission were considered failures and pts who were discharged from the ICU before day 5 were considered successes.

Results: 166 pts were identified, with 202 ICU admissions. The median number of ICU admissions was 1 (1-4), with 138 (84%) having 1 admission, 20 (12%) 2 admissions, 4 (2.4%) had 3 admissions and 3 (2%) 4 admissions respectively. The median length of stay in ICU was 6 days (1-95). The median duration of hospital stay prior to ICU admission was 14 days (0-104). The diagnoses were: AML 28% (n= 57), ALL 8% (n=16), CML 8% (n=16), myelofibrosis 4% (n=7), MDS 4% (n=7), myeloma 11% (n=23), NHL 30% (n=61) and Hodgkin's lymphoma 2% (n=4). Regarding pre ICU treatment, 44% (n=88) received chemotherapy, 11% (n=22) underwent autologous stem cell transplantation and 40% (n=81) allogeneic stem cell transplantation. Of those, 30% had myeloablative and 70% reduced intensity conditioning and 29 (35%) were from HLA identical sibling, 47 (58%) unrelated and 6 (7%) haplo-identical donors. The disease status was complete remission (n=77, 38%), partial remission (n=28, 14%) and stable disease (n=96, 48%).

The reason for admission to ICU was respiratory failure in 53% (n=107), 19% sepsis (n=39), 16% renal failure (n=32) and 11% hemodynamic failure (n=22). The median APACHE II score was 24 (10-51), the median SOFA score was 10 (2-21) and the median SAPS-II score was 45 (0-100). APACHE II and SOFA scores were significantly greater in non-survivors vs survivors (p<0.0001). Overall 101(50%) pts survived their ICU admission and were discharged to the hematology ward. Of these, 31 (30%) died in hospital and 70 (70%) were discharged home. Estimated overall survival was 15% (95% CI 10-23) at 3 years post ICU admission.

For the 5-day ICU trial we selected 138 pts with one admission. The distribution according to the different groups was: a) 56; b) 34; c) 17; d) 17 and e) 14. Overall 58 (42%) successfully passed the trial: a) 30 (53%); b) 14 (41%); c) 4 (23%); d) 7 (41%) and e) 3 (21%). Overall 41 (30%) pts failed the trial and were alive on day 5 and 39 (28%) died before day 5. The overall survival (Figure 1) for the 58 pts who passed the trial was 28% at 3 years. The overall mortality in ICU was 33% (19/58) for those who successfully passed the 5-day ICU trial, and was 71% (29/41) for those who failed. The overall survival for pts that successfully completed the 5-day ICU trial and were discharged to the hematology ward (n=39), was 49% at 3 years.

Conclusions: In this study, 50% of onco-hematologic patients survived their ICU admission. The long-term overall survival was 15% at 3 years. Patients could be stratified according to the reason for admission and given an individualised 5-day trial: those who successfully completed their trial (42%) had a low ICU mortality (33%) and those who were subsequently discharged home had a long-term survival of 49% at 3 years. This study raises the possibility of offering a short-term ICU trial to onco-hematologic patients and perhaps allows for the ceiling of intensive care for those who fail the trial.

Figure

View largeDownload slide

Figure

View largeDownload slide

Close modal