Kohara, H., Ueoka, H., Takeyama, H., Murakami, T., & Morita, T. (2005). Sedation for terminally ill patients with cancer with uncontrollable physical distress. Journal of Palliative Medicine, 8(1), 20-25.

The objective of this study was to investigate the influence on consciousness of sedative drugs to relieve severe physical distress refractory to standard interventions.

The study was a retrospective review of medical records of 124 consecutive patients admitted to a single palliative care unit between January and December 1999 to evaluate the use of sedation, defined as “a medical procedure to palliate patient symptoms refractory to standard therapy by intentionally dimming consciousness.\" Nocturnal sedation was excluded.

• The study reported on a sample of 63 participants.
• The mean age was 64 years, with a range of 35–87 years.
• Of the sample, 67% were males and 33% were females.
• A key disease characteristic was cancer, with the primary tumor sites being lung (38, 60%) and stomach (7, 11%).
• Symptoms requiring sedation were dyspnea (63%), general malaise/restlessness (40%), pain (25%), agitation (21%), and nausea and vomiting (6%).
• Thirty-five patients (54%) had more than one uncontrollable symptom.

This single-site study was conducted in an inpatient setting in Japan.

• Patients were undergoing end-of-life care.
• The study has clinical applicability for palliative care.

• Retrospective chart review was conducted to obtain patient characteristics, symptoms, and treatments.
• Statistical analysis of patient data was performed using unpaired t-test.

• Equivalent daily dose (MEDD) of parenteral morphine
• Parenteral equivalent of midazolam (PME)
• Palliative Performance Scale (PPS) (0 = death, 100 = normal)
• Communication Capacity Scale to measure level of consciousness (0 = awake with no drowsiness, 3 = aroused by verbal stimuli, 5 = unarousable by physical stimuli)

• Palliative Performance Scale results were poor in the group before sedation (10 [46%], 20[37%], 30[14%], 40[3%]).
• The longest duration of sedation was 11 days.
• Midazolam was the most frequently administered sedative (98%) by continuous IV infusion (60%) and SQ infusion (35%).
• Between patients receiving sedation and those not requiring sedation, no significant difference was seen in admission duration or level of consciousness day six–day three before death.
• A significant difference in mean CCS only lasted three days prior to death (day two before death CCS 3.3 [SD = 0.8] [p < 0.05], day one before death CCS 3.8 [SD = 0.7] [p < 0.05], day of death CCS 4.7 [SD = 0.6] [p < 0.01]).
• The mean dose of opioid used was greater in the sedated group but only statistically significantly on day two before death (p = 0.04), day one before death (p = 0.03), and day of death (p = 0.08).

This study contributes descriptive information about the use of terminal sedation (midazolam and opioids) for symptom control and the influence sedation has on the level of consciousness during the last days of life. In this study, patients receiving sedation were significantly drowsier and less responsive only during the last three days of life. What is not known from this study, although it is implied, is the degree of symptom control achieved by this intervention.

Limitations of this study included

• Small sample size of less than 100
• Risk of bias (no control group)
• Risk of bias (no blinding)
• Risk of bias (no random assignment)
• Risk of bias(sample characteristics)
• Selective outcomes reporting
• Questionable measurement validity/reliability
• Nongeneralizable findings
• Limited strength of evidence due to retrospective chart review in single site
• Evidence peripheral to dyspnea relief outcome.

This is helpful, descriptive, and low-level evidence about the use of terminal sedation to control symptoms. No measurement of dyspnea relief was included in the report, although it implies that sedated patients were not in distress.