5323 - 30-Day Mortality in Palliative Radiotherapy

Background

In terminal cancer patients with a limited prognosis, excessive radiotherapy (RT) fractionation can be considered an avoidable harm and poor use of healthcare resource. 30-day mortality (30DM) as a clinical indicator for the avoidance of harm, through appropriate patient selection, in palliative RT has not previously been demonstrated in Hong Kong. We aimed to assess 30DM following palliative RT at our centre.

Methods

A retrospective analysis of 30DM for all palliative RT delivered in Tuen Mun Hospital in 2017. Episodes were identified using the in-house radiotherapy electronic record with linkage to electronic patient records. 30DM was assessed across all patient groups and by sites and fractionation pattern.

Results

1,065 palliative RT episodes were identified. Overall, 199 (18.7%) treatment were delivered to the spine, 151 (14.2%) to the chest, 97 (9.1%) to the brain, and 77 (7.2%) to the bones. Median treatment duration was 5 days (range 1–12). 84.3% received 5 fractions or less. Overall 30DM was 17.7% (188 patients), among which 61 (32.4%) treatment episodes were delivered to female. The most commonly treated diagnosis was lung cancer (39.4%). Factors having a significant impact upon 30DM were sex (male vs female odds ratio [OR] 1.56, 95% confidence interval [CI] 1.12, 2.18) and primary cancer diagnosis (Lip, oral cavity, and pharynx OR 0.31 [0.11, 0.87], genitourinary organs OR 0.57 [0.33, 0.98], and respiratory and intrathoracic organs 1.60 [1.17, 2.20]).Table:

1602P odd ratios for different factors relating to the 30DM of palliative RT

Abbreviation: CI: confidence interval

Conclusions

This is the first large-scale study of 30-day mortality for unselected adult palliative RT in Hong Kong. A balance needs to be achieved between duration of RT course and potential benefits through appropriate patient selection. The observed differences in early mortality by fractionation support the use of this measure in assessing clinical decision making in palliative RT, and further studies in other centres and health care systems are required.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

Shing Fung Lee.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.