Background In end-of-life care, symptoms of discomfort are mainly managed by drug therapy, the guidelines for which are mainly based on expert opinions. A few papers have inventoried drug prescriptions in palliative care settings, but none has reported the frequency of use in combination with doses and route of administration. Objective To describe doses and routes of administration of the most frequently used drugs at admission and at day of death. Setting Palliative care centre in the Netherlands. Method In this retrospective cohort study, prescription data of deceased patients were extracted from the electronic medical records. Main outcome measure Doses, frequency and route of administration of prescribed drugs Results All regular medication prescriptions of 208 patients, 89 % of whom had advanced cancer, were reviewed. The three most prescribed drugs were morphine, midazolam and haloperidol, to 21, 11 and 23 % of patients at admission, respectively. At the day of death these percentages had increased to 87, 58 and 50 %, respectively. Doses of these three drugs at the day of death were statistically significantly higher than at admission. The oral route of administration was used in 89 % of patients at admission versus subcutaneous in 94 % at the day of death. Conclusions Nearing the end of life, patients in this palliative care centre receive discomfort-relieving drugs mainly via the subcutaneous route. However, most of these drugs are unlicensed for this specific application and guidelines are based on low level of evidence. Thus, there is every reason for more clinical research on drug use in palliative care.Electronic supplementary materialThe online version of this article (doi:10.1007/s11096-015-0094-3) contains supplementary material, which is available to authorized users.

Aims Midazolam is the drug of choice for palliative sedation and is titrated to achieve the desired level of sedation. Because of large inter‐individual variability (IIV), however, the time it takes to achieve adequate sedation varies widely. It would therefore greatly improve clinical care if an individualized dose could be determined beforehand. To find clinically relevant parameters for dose individualization, we performed a pharmacokinetic study on midazolam, 1OH‐midazolam (1‐OH‐M) and 1OH‐midazolam‐glucuronide (1‐OH‐MG) in terminally ill patients. Methods Using nonlinear mixed effects modelling (NONMEM 7.2), a population pharmacokinetic analysis was conducted with 192 samples from 45 terminally ill patients who received midazolam either orally or subcutaneously. The covariates analysed were patient characteristics, co‐medication and blood chemistry levels. Results The data were accurately described by a one compartment model for midazolam, 1‐OH‐M and 1‐OH‐MG. The population mean estimates for midazolam, 1‐OH‐M and 1‐OH‐MG clearance were 8.4 l h−1 (RSE 9%, IIV 49%), 45.4 l h−1 (RSE 12%, IIV 60.5%) and 5.1 l h−1 (RSE 11%, IIV 49.9%), respectively. 1‐OH‐MG clearance was correlated with the estimated glomular filtration rate (eGFR) explaining 28.4% of the IIV in 1‐OH‐MG clearance. In addition, low albumin levels were associated with decreased midazolam clearance, explaining 18.2% of the IIV. Conclusion Our study indicates albumin levels and eGFR as relevant clinical parameters to optimize midazolam dosing in terminally ill patients. The correlation between low albumin levels and decreased midazolam clearance is probably a result of inflammatory response as high CRP levels were correlated in a similar way.

This is one of the first validation studies in which BIS monitoring in end-of-life patients is described. BIS monitoring is feasible in unconscious terminally ill patients. However, based on our results, the wide range of BIS values in deeply sedated and comfortable patients seems to hamper its use in daily clinical practice.