gms | German Medical Science

Objective: In the quality analysis of acute brain diseases such as stroke, subarachnoid hemorrhage or traumatic brain injury mortality often is given as an indicator of treatment quality. Against the background of a constantly growing self-determination, the number of patients with living wills is increasing and subsequently, a decision to withdraw life-sustaining treatment is made in cooperation with the relatives. In this context, the question arises to what extent death can be evaluated as a suitable quality parameter for possible therapeutic success. The aim of this retrospective monocentric study was to investigate the number of deaths as a result of treatment de-escalation with regard to neurosurgical diseases.

Methods: The cases of death in neurosurgical patients in the intensive care unit (ICU) were analysed in the period between January 2017 to June 2019. The first step was to investigate how many patients died during the study period and how large their share was in relation to the total disease specific number of neurosurgical patients treated in the ICU. In a further step an analysis was performed with regard to the main diagnoses intracerebral hemorrhage (ICH), subarachnoid hemorrhage (SAH) and traumatic brain injury (TBI) over the same period. The individual course of treatment was worked out with regard to a possible therapy de-escalation. Finally, an investigation was carried out to determine when a therapy de-escalation occurred.

Results: A total of 1531 ICU patients (608 patients with ICH, 220 patients with SAH, 703 patients with TBI) were included in our analysis. Overall 176 patients (11,5 %) died during time course. Mortality rates were 14,3 % for ICH, 13,2 % for SAH and 8,5 % for TBI. The individual course of treatment was worked through for the patients who had passed away and finally a division into five different groups or decision paths was made with regard to a possible therapy de-escalation. In 88 cases (50 %), the treatment was de-escalated during the course, with the time of de-escalation varying from day 0 to day 71. 60 % of patients with ICH, 34 % with SAH and 42 % with TBI had their treatment de-escalated. In almost one fifth of these cases, death could have been prevented from a medical point of view.

Conclusion: Mortality as an indicator of treatment quality only appears to be appropriate if the proportion of patients with therapy de-escalation is taken into account.