175. Cognitive impairments prior to spine surgery: association with symptoms severity and impact on short-term outcome

Abstract

<h3>BACKGROUND CONTEXT</h3> Previous studies have shown that cognitive impairment within the surgical population is associated with prolonged hospital stays, increased postoperative complications, and even death. While studies have examined the role that cognitive impairment may have on the incidence of postoperative delirium, their role in disease activity and recovery following surgery are not well understood. <h3>PURPOSE</h3> Our objectives were to determine the 1) prevalence of preoperative cognitive impairment (CI) among patients presenting for spine surgery at a large academic medical center; 2) effect of preoperative CI on symptom burden and health related quality of life (HRQOL); and 3) effect of preoperative CI on short-term HRQOL. We hypothesized that preoperative CI would mirror the observed age-adjusted rates in the general population; patients with preoperative CI would endorse more severe pain and pain-related disability and lower HRQOL; and patients with preoperative CI would experience less improvement in HRQOL following spine surgery. <h3>STUDY DESIGN/SETTING</h3> The current study is a prospective cohort study of patients who presented for spine surgery at an academic medical center. <h3>PATIENT SAMPLE</h3> Eligible patients were English-speaking and aged 18 years or older who were scheduled for elective surgery for degenerative lumbar or cervical spine disease or for correction of spinal deformity. We report on the experience of 436 patients. Participants' mean age was 57 years (standard deviation, 18 years) who were evenly split by gender (52% female) and most were non-Hispanic (95%), and white (81%). All participants were treated at the same center by 5 fellowship-trained, board-certified spine surgeons with 6–26 years in practice. <h3>OUTCOME MEASURES</h3> Participants completed an assessment before surgery of sociodemographic and clinical information. Sociodemographic information comprised age, sex, race/ethnicity, highest educational attainment, annual household income and relationship status. Highest educational attainment was categorized as less than 4-year college degree; 4-year college degree; or post-graduate degree. Annual household income was categorized as <$30,000; $30,000–$80,000; or >$80,000. Relationship status was used to determine whether a participant lived alone or with a partner. Clinical information was presence of comorbid conditions and use of opioid medication within the past 30 days. Participants completed assessments of their health and wellness before and after surgery. After surgery, assessments were completed during the immediate postoperative period (6–12 weeks) and at 6, 12, and 24 months. These assessments were the Pain Numeric Rating Scale (NRS) for back/leg and neck/arm, the Oswestry Disability Index (ODI) or the Neck Disability Index (NDI) and the Patient-Reported Outcome Measurement Information System (PROMIS-29) health domains. <h3>Methods</h3> Using preoperative score on the PROMIS Cognitive Abilities Short Form, we grouped participants as having mild (score < 40) and moderate (score < 30) preoperative CI. These scores corresponded to 1 and 2 SD below the age- and gender-adjusted mean score. We compared the observed prevalence of preoperative CI to the age- and gender-adjusted population rate using a one-sample Z proportion test. We compared pain and pain-related disability and HRQOL between those with and without preoperative CI using t-test. General linear models were fit (adjusting for age, sex, comorbid condition, and opioid use within 30 days before surgery) to estimate the influence of pre-operative CI on recovery following surgery. Alpha = 0.05. <h3>Results</h3> Preoperative CI was reported as mild by 83 participants (19%) and moderate by 19 (4%). This was not significantly greater than age- and gender-adjusted rates (p>.05). Participants who reported mild preoperative CI experienced significantly more back pain (8.3 vs 7.3, p=.011) but no difference in leg, neck or arm pain; greater pain-related disability (ODI: 56 vs 41, p<.001; NDI: 48 vs 36, p<.001). In addition, these participants reported worse HRQOL (PROMIS pain interference: 68 vs 64, p<.001; physical function: 34 vs 37, p<.001; fatigue: 61 vs 52, p<.001; anxiety: 58 vs 51, p<.001; depression: 57 vs 49, p<.001; sleep disturbance: 60 vs 54, p<.001; and satisfaction with social roles: 38 vs 43, p<.001). Participants who reported moderate preoperative CI experienced similar though more pronounced differences in pain and pain-related disability and HRQOL. Following surgery, mild preoperative CI was associated with less recovery of HRQOL when examine the linear slope of change in score on PROMIS pain interference: -0.4 vs -1.0, p=.001). Moderate CI was associated with less recovery in HRQOL (PROMIS pain interference: -0.1 vs -0.9, p=.045; fatigue: +0.58 vs -0.4, p=.03; and anxiety: +0.9 vs -0.2, p=.012). <h3>Conclusions</h3> About one in five spine surgery patients reported preoperative CI. These individuals experience greater burden in terms of increased pain and pain-related disability and worse HRQOL. Pain interference on activities of daily living persist in these patients following surgery. These results highlight the importance of addressing patients' cognitive difficulties both before and after surgery. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.