Factors contributing to memory of acute pain in elderly patients undergoing planned and unplanned hip surgery

Background:
Retrospective evaluations of pain not only serve as the basis for medical diagnoses, choice of treatment, and proof of symptom improvement, but they also may affect subsequent pain experience and the willingness to return for painful procedures, and they may even contribute to the development of chronic pain. However, consistent evidence whether recalled pain is accurate, overestimated, or underestimated, is lacking. Among factors that can influence pain memory, such as time, actual or expected pain intensity, anxiety, negative and positive affect, also the context of the painful experience has been associated with over- or underestimation of recalled pain. Furthermore, mood dependent memory, occurring when the congruence between affective states at encoding and retrieval facilitates memory, might account for some inconsistencies in the existing pain memory research. Though several studies have explored factors that potentially influence pain memory, few have attempted to examine broad affective and contextual contributions to this phenomenon.
Aims:
1) To determine the accuracy of postoperative pain recall after a 3-month delay, with respect to the context of the painful event and the congruence of affective states during the initial experience and its recall.
2) To identify potential sensory, cognitive and affective predictors of remembered pain intensity and unpleasantness.
Methods:
This prospective, observational study investigated two groups of elderly patients undergoing painful medical procedures in a positive or negative context, i.e. planned (N = 40; mean age = 71.25+/-6.42; 22 female) or unplanned hip surgery (N = 31; mean age = 78.9+/-9.76; 25 female). Data was collected 24-48 hours pre-surgery (T1) and post-surgery (T2), and 3-months after the procedure (T3). At each time point, participants provided the ratings of experienced pain intensity and unpleasantness and state anxiety on Numeric Rating Scales (NRS). Positive and negative affect was measured with Positive and Negative Affect Schedule (PANAS). Participants reported expected pain intensity and unpleasantness (NRS) and underwent cognitive assessment with Mini-Mental State Examination (MMSE) in the pre-surgical period only. Recalled pain intensity and unpleasantness ratings (NRS) were collected only after 3-month delay, and until that point the participants were not aware that their memory of pain will be tested.
Results:
Repeated-measures analyses of covariance (ANCOVAs, with group age and MMSE scores as covariates) followed by planned comparisons (experienced vs recalled measures of postoperative pain intensity and unpleasantness; T2 vs T3 measures of positive and negative affect) revealed no significant interactions and differences. The results suggest that memory of pain was accurate and mood states in the postoperative period and after the delay were congruent, regardless of the context of the surgery, indicating a possible mood dependence effect on pain memory. Stepwise forward multiple regression analyses found that cognitive functioning and T2 experienced pain accounted for significant proportion of the variance in remembered pain intensity (37%) and, together with T3 negative affect and T1 positive affect, also in pain unpleasantness (59%) after planned surgery. Age, T1 anxiety, T2 negative affect and T3 positive affect were found to be significant predictors of remembered pain intensity (52%) and unpleasantness (47%) after unplanned surgery.
Conclusion:
Memory of acute postoperative pain appears to remain accurate after long delay, which may be partly attributable to mood dependent memory. Furthermore, a variety of factors influence pain memory, depending on the context of painful experience. Planned surgery appears to be associated mostly with sensory predictors of remembered pain, whereas memory of pain after unplanned surgery seems to depend not on the actual sensory experience, but on affective factors (especially anxiety), possibly due to the largely negative context of acute trauma. Study findings have implications for future research on pain memory and may facilitate improvements in perioperative hospital care.