ABSTRACT
Patients with haematological conditions can become critically ill due to their underlying disease or secondary to treatment complications. The BSH published guidelines on the management and admission to intensive care unit (ICU) of critically ill adult patients with haematological malignancies in 2015 (Wise et al, BJHaem) Here we present the first published data on compliance with this guideline at the University Hospital of Wales, Cardiff. We performed a retrospective study of 30 haematology patients referred to ICU between September 2020 and July 2021. Patients were identified from an intensive care database and individual patient data were collected using patients medical records. The BSH audit template was used to evaluate the patient pathway. Additional data such as performance status, advanced care planning were assessed and are shown in the table below. All were monitored by NEWS score throughout admission. Referral to ICU was made by haematology registrars in 42%, medical registrars in 26% and haematology consultants in 32%. Sixteen (53%) patients had ICU reviews as part of planned discussions which occurred at a median of 23 (2-243) h of becoming unwell, whereas 14 (47%) patients were reviewed by ICU at the time of a crash call. Importantly, the majority (66%) of crash calls occurred outside normal working hours highlighting the importance of adequate out of hours (OOH) medical cover. With regards to communication, 68% of patients admitted to ICU had documentation with either patient or relative at the time of ICU admission. Of the patients referred, 19 (63%) were accepted for admission to the ICU. The average length of hospital stay prior to ICU admission was 17 days and the average length of stay in ICU was 5 (1-58) days. The majority of patients (68%) had sepsis and 14 (74%) required invasive ventilation. Eight cases were neutropenic and had a higher death rate of 75% compared with 55.6% of non-neutropenic patients. Overall, mortality rate during ICU stay was 68%. By contrast, in eight patients who were initially deferred but later accepted, the mortality rate was 75%. These patients were accepted for transfer to ICU typically 2-3 days after the initial deferral. Haematology inputs were documented in 89% of patients during the ICU stay. In summary, clear documentation of escalation plans and resuscitation status is essential to enabling prompt treatment decisions in a deteriorating patient. Our audit highlights the need for improvement in this area and we take lessons from medical admissions during the COVID pandemic where all patients have a clear escalation plan. The data also highlighted the majority of crash calls occurred OOH. A recent introduction of a remodified critical care outreach service, PART (patient at risk team) will be crucial for service development and coordination of care between ICU and haematology. In a COVID era regular ICU and haematology communication may be challenging thus virtual mortality and morbidity meetings can be helpful. Communication with patients and families is encouraged at all phases of care with more emphasis around the ICU admission. A re-audit is planned in 1 year to evaluate practice which will evaluate the additional benefit of the PART team in co-ordination of care. (Table Presented).