Introduction: Despite the prevalence of significant alcohol use, the effect of alcohol consumption on patients undergoing total knee arthroplasty (TKA) is poorly understood. Other forms of substance abuse, such as narcotic dependence, decrease patient pain tolerance and increase self-reported pain scores, while excessive alcohol consumption has been associated with adverse outcomes following other elective surgical procedures. In the context of TKA, perception of pain is also likely linked to the ability to achieve functional postoperative range of motion (ROM), one of the most important predictive indicators of patient satisfaction. It is not well understood whether heavy alcohol users who are TKA candidates come to surgery with higher degrees of osteoarthritis (OA) and/or higher levels of self-reported pain nor do we clearly understand the effect on postoperative ROM. The purpose of this study was thus to evaluate how levels of alcohol consumption affect: 1) The severity of preoperative knee OA, as measured by radiographic classification and visual analog scores (VAS) of pain and; 2) Pre- and post-TKA ROM. We hypothesized that preoperative TKA patients who drink more alcohol would report more pain, have more severe arthritis, and have worse postoperative ROM compared to rare or non-drinkers.
Methods: An Institutional Review Board (IRB) approved retrospective cohort study of TKA patients from a single surgeon at a large US academic center over the previous 10-year period was performed. Patients were identified through International Classification of Disease (ICD) query and the electronic medical record data were tabulated and reviewed by another author who was not the surgeon. Alcohol consumption was self-reported by patients at their preoperative appointment and they were categorized by drinks/week as: 1) Abstainers; 2) Occasional drinkers (≤ 1); 3) Moderate drinkers (1 to 7), and; 4) Heavy drinkers (≥ 7). Pain scores were self-reported on a 0-10 VAS at the preoperative appointment. ROM was recorded at the preoperative visit; postoperative ROM was recorded at 2 weeks, 6 weeks, 6 months, and 12 months. ROM was measured as the degrees of knee extension subtracted from the degrees of knee flexion, which were measured using a goniometer . The surgeon blindly evaluated radiographs of all patients using the Kellgren-Lawrence (0-4) classification. Patients were excluded if they did not report alcohol intake, did not have a recorded preoperative baseline ROM, or did not have adequate follow-up for ROM measurements. Statistical analyses comprised regression modeling with pairwise post-hoc t-tests.
Results: A total of 764 TKAs (316 male, 448 female) were identified and analyzed in this study, with the following breakdowns by alcohol consumption group and gender: Abstainer = 302 (92M, 210F); Occasional = 260 (100M, 160F); Moderate = 140 (88M, 52F); Heavy = 62 (36M, 26F). Pain scores in Heavy use males were lower compared to Moderate use males (p<0.03), Occasional use males (p<0 .03), and Abstainer males (p<0.01) . Pain scores in Moderate use females were lower compared to Abstainer females (p<0.01) . Other group comparisons were not found to be significantly different. After blinded review of the preoperative radiographs, no correlations were found between alcohol consumption group and Kellgren-Lawrence radiographic classification.
After exclusion of patients without adequate follow-up, a total of 499 TKAs (263 male, 236 female) were included in the pre and postoperative ROM analysis, with the following breakdowns by alcohol consumption group and gender: Abstainer = 159 (77M, 82F); Occasional = 156 (79M, 77F); Moderate = 123 (71M, 52F); Heavy = 61 (36M, 25F). ROM by alcohol consumption group at 12 months in females was as follows: Abstainer = 117.8 ± 11.4; Occasional = 114.9 ± 11.6; Moderate = 115.6 ± 12.2; Heavy = 124.5 ± 4.4. ROM by alcohol consumption group at 12 months in males was as follows: Abstainer = 118.3 ± 11.8; Occasional = 118.0 ± 10.5; Moderate = 117.4 ± 11.1; Heavy = 119.0 ± 7.4. Within both populations, our analyses found no statistically significant differences in ROM following TKA between alcohol consumption groups.
Conclusion: Increased self-reported alcohol consumption in both genders prior to TKA was associated with lower preoperative pain scores, similar degrees of radiographic OA, while it did not affect postoperative ROM, rejecting our hypotheses. To our knowledge this is one of the first studies to assess the effect of chronic alcohol use on self-reported pain and postoperative ROM in TKA patients. Our results are clinically important because they suggest that heavy alcohol use does not adversely affect the preoperative pain perception nor postoperative ROM. The patients that reported higher alcohol consumption in our cohort came to TKA surgery with lower VAS pain scores, similar degrees of OA, and achieved similar ROM postoperatively when compared to less active alcohol users. This is in contrast to narcotic dependency, which historically correlates with higher perceptions of pain causing poorer postoperative ROM thus leading to worse patient satisfaction. The results of decreased pain in heavier drinkers may reflect the analgesic effects of alcohol on the perception of pain in OA patients. While our results show no relationship between alcohol consumption and radiographic OA severity, they are limited to a snapshot at one point in time and further investigation is needed into the effects of alcohol consumption on OA progression over a long period of time. Further studies are warranted to investigate the effects of alcohol consumption on other TKA outcomes such as pain medication use, restoration of function, and patient satisfaction.
