Frailty status as a potential factor in increased postoperative opioid use in older adults

Abstract

Background: Prescription opioids are commonly used for postoperative pain relief in older adults, but have the potential for misuse. Both opioid side effects and uncontrolled pain have detrimental impacts. Frailty syndrome (reduced reserve in response to stressors), pain, and chronic opioid consumption are all complex phenomena that impair function, nutrition, psychologic well-being, and increase mortality, but links among these conditions in the acute postoperative setting have not been described. This study seeks to understand the relationship between frailty and patterns of postoperative opioid consumption in older adults.

Methods: Patients ≥ 65 years undergoing elective surgery with a planned hospital stay of at least one postoperative day were recruited for this cohort study at pre-anesthesia clinic visits. Preoperatively, frailty was assessed by Edmonton Frailty and Clinical Frailty Scales, pain was assessed by Visual Analog and Pain Catastrophizing Scales, and opioid consumption was recorded. On the day of surgery and subsequent hospitalization days, average pain ratings and total opioid consumption were recorded daily. Seven days after hospital discharge, patients were interviewed using uniform questionnaires to measure opioid prescription use and pain rating.

Results: One hundred seventeen patients (age 73.0 (IQR 67.0, 77.0), 64 % male), were evaluated preoperatively and 90 completed one-week post discharge follow-up. Preoperatively, patients with frailty were more likely than patients without frailty to use opioids (46.2 % vs. 20.9 %, p = 0.01). Doses of opioids prescribed at hospital discharge and the prescribed morphine milligram equivalents (MME) at discharge did not differ between groups. Seven days after discharge, the cumulative MME used were similar between cohorts. However, patients with frailty used a larger fraction of opioids prescribed to them (96.7 % (31.3, 100.0) vs. 25.0 % (0.0, 83.3), p = 0.007) and were more likely (OR 3.7, 95 % CI 1.13-12.13) to use 50 % and greater of opioids prescribed to them. Patients with frailty had higher pain scores before surgery and seven days after discharge compared to patients without frailty.

Conclusions: Patterns of postoperative opioid use after discharge were different between patients with and without frailty. Patients with frailty tended to use almost all the opioids prescribed while patients without frailty tended to use almost none of the opioids prescribed.

Fig. 1

Time course of the study. Patients were interviewed for a preoperative evaluation during their pre-anesthesia clinic appointment. Preoperative evaluation included Visual Analog Scale pain rating, Clinical Frailty Scale and Edmonton Frailty Scale assessments, recorded MME of opioid use, and Pain Catastrophizing Scale assessment for pain catastrophizing. All patients were hospitalized after surgery for at least one postoperative day. On each day of the postoperative hospital stay, pain ratings were obtained from nurses' medical administration records or electronic medical records as a Numerical Rating Score or Verbal Rating Scale, and opioid use was captured in MME from the electronic medical record. Seven days after discharge from the hospital, patients reported pain (using the Numerical Rating Score) and opioid use (in MME) during a phone or email interview. Abbreviations: CFS, Clinical Frailty Scale; EFS, Edmonton Frailty Scale; VAS, Visual Analog Scale; PCS, Pain Catastrophizing Scale; MME, morphine milligram equivalents; NRS, Numerical Rating Score; VRS, Verbal Rating Scale

Fig. 2

Recruitment and follow-up of patients screened during their visit to the pre-anesthesia clinic. One hundred and twenty-seven patients that presented at the pre-anesthesia clinic during the study period were eligible and screened for participation. A pre-anesthesia evaluation was completed for 117 patients, and 102 patients were followed through their surgery and hospital stay. Follow-up seven days after hospital discharge was completed in 90 patients

Fig. 3

MME median (IQR) usage for patients with and without frailty. Preoperative assessment of MME was performed at the pre-anesthesia clinic. Hospitalization Day 0 is the day of surgery. Hospitalization ranged from 1 to 26 days and assessment was collected from the electronic medical record. The number of patients is reduced in consecutive days because patients were discharged from the hospital. The number of patients who stayed more than 3 days after surgery was small and therefore not presented. Assessment after hospital discharge was conducted by patient self-report. There was no significant difference in the median amount of opioids used during hospitalization or after hospital discharge between patients with and without frailty

Fig. 4

Percent of MME of opioid prescription consumed seven days after hospital discharge. Patients were interviewed by phone or email seven days after hospital discharge to determine how much of their opioid prescription was used. Opioid usage and prescription were recorded in MME and the usage was converted to a percent of the total prescribed opioids. Patients without frailty were more likely to use less than 20 % of the opioid prescription (47.3 % vs. 18.8 %, p = 0.036). Patients with frailty were more likely to use 50 % and greater of the opioid prescription (68.8 % with frailty vs. 36.5 % without frailty, p = 0.018). *p < 0.05

Fig. 5

Pain ratings for frail and non-frail patients using comparable 11-point pain scales. Preoperative assessment was done at the pre-anesthesia clinic using the visual analog scale. Hospitalization ranged from 1 to 26 days and assessment was done using numerical rating score and verbal rating score. Assessment after hospital discharge used numerical rating score. Hospitalization Day 0 pain scores are after surgery on the day of surgery. Patients with frailty had more pain 7 days after hospital discharge (6.5 vs. 3.0, p < 0.05) compared to patients without frailty, but patients had no differences in pain scores on the day of surgery or during hospitalization. *p < 0.05