The present study confirmed the hypothesis that lower preoperative MMSE scores are associated with a higher risk of developing delirium after major surgery in hemodialysis patients. The study also showed that two of the categories, orientation, and language and praxis, among the five categories in the MMSE are stronger predictors of postoperative delirium compared to the other three categories. Although it did not have a very high accuracy, ROC analysis showed an AUC of 0.740, which is acceptable. Several previous reports have shown that low preoperative MMSE scores are associated with a higher risk of postoperative delirium5–8. However, data on the correlation between the two in hemodialysis patients was so far lacking. This is the first study to demonstrate this point, including the relative effects of the various categories of the MMSE.
The average MMSE score was previously reported as 27.0 points in a community-based cohort survey in China, which included a total of 11,732 non-CKD participants older than 65 years18. Several studies have reported MMSE scores in CKD patients. The average MMSE score was 26.4 points in a study of 112 CKD patients older than 65 years with a mean eGFR of 24.0 mL/min/1.73m2 19. Average MMSE score was 24.1 in a study of 75 hemodialysis patients with a mean age of 59.2 years20. The average MMSE score in our study of 47 hemodialysis patients with a mean age of 74.7 years was 25.5. Although the data of MMSE scores in CKD patients, including dialysis patients, are very limited, these data, including ours, suggest that the presence of CKD and its severity might affect MMSE score.
Among the five item categories in the MMSE, namely orientation, registration, attention and calculation, recall, language and praxis, only orientation and language and praxis were significantly associated with postoperative delirium. By investigating 100 aged subjects (average age, 71.5 years old), Solfrizzi et al. showed that aging was associated with reduced scores for the MMSE items of “orientation” and “recall” 21. Ralat showed that having a psychiatric disorder, including bipolar disorder, major depressive disorder and schizoaffective disorder, was associated with a lower MMSE score for the item “orientation” 22. More importantly, Franco et al. showed that lower scores for the MMSE item “orientation” was associated with the development of hypoactive delirium23. However, which MMSE item has a stronger effect on the development of postoperative delirium has not been examined. Additionally, although we identified “language and praxis” as a predictive factor for postoperative delirium, it is unclear whether this result is distinctive for patients with kidney dysfunction. A meta-analysis including 42 studies showed that language skill was worse in hemodialysis patients24. Another review article commented that the most commonly impaired cognitive function in dialysis patients is executive function, which cannot be examined by MMSE25. CKD patients have multifactorial and characteristic risk factors, including cerebrovascular disease, renal anemia, secondary hyperparathyroidism, dialysis disequilibrium, and uremic toxins26. Therefore, analysis specific to various pathologies will be needed in future research.
To date, due to the lack of data, we do not know whether having CKD, especially with a need for dialysis, is a risk factor for development of postoperative delirium or not. Adogwa et al. reported that postoperative delirium was significantly higher in CKD group patients9. However, their report did not evaluate the severity of CKD in terms of dialysis dependence9. Davani et al. also demonstrated that the risk of development of delirium after hip fracture surgery was significantly greater in patients with CKD (average serum creatinine level of 1.5 mg/dL) 10. These limited data suggest that CKD is a potential risk factor for postoperative delirium. Generally, older age is a strong risk factor for postoperative delirium. A systematic review including 24 studies demonstrated that increasing age was the most consistent risk factor for postoperative cognitive dysfunction27. With progressive aging of the population, the recent mean age at the induction of dialysis therapy is over 70 years old in Japan28. Hence, this issue will become more and more important in the future, especially in other aging societies such as Japan.
The results of our study indicate the possibility that preoperative screening by MMSE is useful as a predictor of postoperative delirium in hemodialysis patients. This result is important in terms of identifying patients at a higher risk of postoperative delirium, since it might help health care workers to implement preventive interventions29. A systematic review and meta-analysis that included 31 randomized controlled trials concluded that multicomponent interventions, such as decreasing medication use and pain and anxiety, using bispectral index-guided anaesthesia, and antipsychotics and dexmedetomidine treatment can reduce the severity and duration of postoperative delirium in elderly patients29. It is also necessary to confirm whether, along with these interventions, preoperative screening by the MMSE might improve the severity and duration of postoperative delirium in hemodialysis patients.
The limitations of the present study are its small sample size, the fact that it was a single-center study with no control group, and that it lacked data of a hard endpoint such as death. There was also variation in the patients’ postoperative conditions, including whether or not they were admitted to the ICU. Further, we did not evaluate patients’ present medications or anesthetic methodology, both of which are possible contributors to postoperative delirium. Next, the MMSE has not been validated in ESKD patients, and it can be affected by race and ethnicity, as well as language16, 30. In relation to this, there are concerns about the generalizability of our results because the present study included only Japanese patients.
In conclusion, our data suggests that preoperative MMSE score is a predictor of postoperative delirium in hemodialysis patients. The data also indicate that a 1-point decrease in MMSE scores increases the risk of development of postoperative delirium by 31%. The predictive accuracy of preoperative MMSE scores by ROC analysis showed an AUC of 0.740, which is a borderline acceptable value in clinical practice. Further studies with large sample sizes and of higher quality are needed to determine whether preoperative MMSE accurately predicts postoperative delirium in hemodialysis patients.