Statin Use and Delirium Risk: An Updated Systematic Review and Meta-Analysis

Supplemental Digital Content is Available in the Text. Background: Findings on the association of statin use with delirium risk are inconsistent. The Study Question: Is statin use associated with delirium risk? Study Design: We searched PubMed, the Cochrane Library, and the EMBASE database, limiting the search to human patients and articles in English published until December 31, 2021. The effect size and 95% confidence interval (CI) were defined as the odds ratio (OR) and 95% CI, respectively, to indicate the difference in the incidence of delirium between statin use and nonuse groups. A random-effects model was selected in the case of high heterogeneity of study populations. We used funnel plots, Egger test, Duval and Tweedie trim-and-fill approach, and the classic fail-safe N to assess publication bias. Results: Of a total of 264 identified studies, 13 were selected for the qualitative review—4 RCTs and 9 observational cohort studies. Statin use was not associated with low delirium risk (pooled OR, 0·82; 95% CI, 0·64–1·04; P = 0·09). Substantial statistical heterogeneity was observed (I2, 90%). Visual inspection of the funnel plot of ORs from the studies revealed symmetry. Using the Grading of Recommendations Assessment, Development, and Evaluation approach, we assigned the evidence a rating of C and a weak recommendation for this review. Conclusions: Statin use is not associated with delirium risk. More comprehensive RCTs are required to confirm the results.


INTRODUCTION
Delirium is an acute confusional state characterized by disturbance in attention, awareness, or cognition over a short period of time. 1 Approximately 30% of older patients with medical conditions experience delirium at some time during a hospital stay. 2 A relatively high proportion of surgical (12%-20%) 3,4 and critically ill (55.8%) patients experience delirium. 5 Patients with delirium experience prolonged hospitalization and functional and cognitive decline and have higher mortality and institutionalization risks. 6,7 Multicomponent, nonpharmacological interventions that manage many modifiable risk factors seem to reduce its incidence. 8 In addition, complete ABCDEF bundle performance was associated with lower likelihood of delirium (adjusted odds ratio [OR] 0.60; 95% confidence interval [CI], 0.49-0.72) for critically patients. 9 The ABCDEF bundle includes: Assess, Prevent, and Manage Pain, Both Spontaneous Awakening Trials and Spontaneous Breathing Trials, Choice of analgesia and sedation, Delirium: Assess, Prevent, and Manage, Early mobility and Exercise, and Family engagement and empowerment. 10 The available evidence does not support the use of medications to prevent delirium in high-risk settings such as acute care, intensive care, cardiac surgery, and other postoperative care. 11,12 Investigators continue to evaluate the potential benefits of cholinesterase inhibitors, 13 antipsychotic agents, 14 dexmedetomidine, 15 melatonin, 16 and statins. 17 Statins have pleiotropic properties, including antiinflammatory, immunomodulatory, and antithrombotic. 18,19 The pathophysiology of delirium is complex and multifactorial. Inflammatory alteration in the central nervous system is known to be a crucial factor. 20 In 2016, Vallabhajosyula et al included 6 observational studies in a meta-analysis, the results of which indicated that statin use was not associated with a decreased risk of delirium in critically ill and cardiac surgery patients. Recently, inconsistent evidence has been reported, including that from randomized controlled trials (RCTs) [21][22][23][24] and observational studies. [25][26][27] Therefore, we performed an up-to-date systematic review and meta-analysis to elucidate the association between statin use and delirium risk among critically ill and surgical patients.

METHODS
This meta-analysis was conducted strictly in accordance with PRISMA guidelines. 28 This study was prospectively registered in the PROSPERO registry, with the registration number CRD42022307045. The PRISMA checklist is provided in Supplemental Digital Content (see Table S1, http://links.lww.com/AJT/A137).

Search strategy
We included RCTs and observational studies (cohort or case-control studies) that evaluated the effects of statins on delirium in critically ill or surgical patients in an intensive care unit (ICU). We searched PubMed, the Cochrane Library, and the EMBASE database, limiting the search to human patients and articles in English published until December 31, 2021. The details of the search strategy are provided in Box 1. All identified articles were imported into EndNote software (version X8, Thomson ResearchSoft, CT) to remove duplicate records automatically. Two reviewers (Y.H.C. and J.Y.W.) screened all titles and abstracts independently and evaluated relevant articles. In addition, we searched available bibliographies and review articles for additional potential articles that can be included in this study. Disagreements were resolved through consensus and consultation with a third reviewer (M.J.L).

Data extraction and outcome measures
Using a standardized data extraction form, we obtained information on the study year, study location, study population, participant characteristics, delirium outcome, and effect sizes and their 95% CI. For studies that reported only the crude effect size, we contacted authors and requested them to provide data on the adjusted effect size and relevant information. The outcomes of delirium were assessed by using the Confusion Assessment Method (CAM), CAM-ICU, or diagnostic codes from administrative databases. We obtained sufficient published data to calculate the estimates of the crude or adjusted OR with the 95% CI of delirium risk related to statin use in eligible studies. In addition, delirium-free days were counted as an outcome in a few studies.

Risk of bias assessment
To examine the quality of the included studies, we used the Cochrane Risk of Bias tool for RCTs 29 and the Newcastle-Ottawa scale for observational studies Supplemental Digital Content (see Table S2, http:// links.lww.com/AJT/A137). 30 The Cochrane Risk of Bias tool assesses the potential sources of bias including selection bias (random sequence generation and allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessment), attrition bias (incomplete outcome data), and reporting bias (selective reporting). Each study was categorized as having low risk (green), unclear risk (yellow), or high risk (red) of bias. The risk of attrition bias was considered to be low if the dropout rate was lower than 20% due to the inclusion of few participants in each of the studies. For assessing reporting bias, the protocols for each study were searched on clinicaltrials.gov and biomedcentral.com. The Newcastle-Ottawa scale evaluates 3 major sources of bias in epidemiological studies by using 5 items: 1 for selection bias, 2 for measurement bias, and 2 for confounding bias. A study is classified in each domain as having high or low risk in accordance with prespecified criteria. All included studies were evaluated by 2 pharmacists (Y.H.C. and J.Y.W.) independently; any discrepancies were resolved through discussion.

Statistical analysis
We conducted a meta-analysis to determine the association of statin use with delirium in RCTs and observational studies. The effect size and 95% CI were defined as the OR and 95% CI to indicate the difference in the incidence of delirium between statin use and nonuse. A random-effects or fixed-effects model was selected to calculate the pooled effect size, and a random-effects model was used if the study population had high heterogeneity. We assessed statistical heterogeneity across studies using the I 2 statistic. 31 I 2 values of ,50%, 50%-75%, and .75% are deemed to indicate low, moderate, and high heterogeneity, respectively. Exclusion sensitivity analysis was conducted to investigate the effect of individual studies. We performed subgroup analysis on the basis of the study design (RCTs vs. other observational studies), country in which the studies were performed, publication before 2015 versus after 2015 and hydrophilic or lipophilic statins. We used funnel plots, Egger test, Duval and Tweedie trim-and-fill approach, and the classic failsafe N to assess publication bias. 32,33 In addition, we performed meta-regression by adjusting for age and male sex to explore the potential sources of heterogeneity. The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach 34 was employed to evaluate the quality of evidence; in this approach, the 4 grades are high, moderate, low, and very low quality. All statistical analyses were performed using RevMan5 and Comprehensive Meta-Analysis Software.

Search results and study characteristics
A total of 264 database records were screened, and the eligibility of 23 full-text articles was assessed. Eight review articles, 3 duplicate reports, and one report on costeffectiveness were excluded from this meta-analysis. Finally, 13 articles were selected for the qualitative review: 4 RCTs and 9 observational cohort studies ( Figure 1). Table 1 summarizes the characteristics of the 13 included studies. The studies included 300,590 participants from 8 countries. The mean age of the study participants ranged from 51$0 to 73$9 years across the studies. The percentage of male patients ranged from 7$4%-88%.

Quality assessment
Supplemental Digital Content (see Figure S1, http:// links.lww.com/AJT/A137) presents the results of the quality assessment for the 4 RCTs. Three RCTs 21,22,24 were discovered to have low risk in all the domains of assessment; only one study, which was conducted by Sohrevadi et al, 23 had an unclear risk of random sequence generation, selective-reporting, and other biases. Supplemental Digital Content (see Table S2, http://links.lww.com/AJT/A137) summarizes the results of the quality assessment for the 9 observational cohort studies. The quality assessment scores ranged from 6 to 8. Five studies did not demonstrate that outcome of interest was not present at start of study. None of the studies described adequacy of follow-up of cohorts. All the 9 studies adjusted for at least one of the other potentially crucial confounders (see Table S3, Supplemental Digital Content, http:// links.lww.com/AJT/A137). Thus, all the studies were evaluated to be of high quality.

Meta-analysis, sensitivity analysis, and publication bias
Overall, statin use was not associated with a low risk of delirium (pooled OR, 0$82; 95% CI, 0$64 to 1$04; P 5 0$09; Figure 2). In addition, substantial statistical heterogeneity was noted (I 2 , 90%). When we removed any 1 of the 11 studies at one time, high heterogeneity (I 2 , 75%-91%) was still observed. When the study by Redelmeier et al 35 was removed, the heterogeneity was 75%. In the sensitivity analysis, the results were inconsistent after each study was individually removed (see Figure S2, Supplemental Digital Content, http:// links.lww.com/AJT/A137). The association between statin use and delirium risk became significant when we removed the study conducted by Redelmeier et al (OR, 0$76; 95% CI, 0$62-0$93; P 5 0$009) or that by Mariscalco et al (OR, 0$77; 95% CI, 0$60-0$98; P 5 0$04). 35,36 Visual inspection of the funnel plot of ORs from these studies revealed symmetry (Figure 3). The result of Egger regression intercept approach indicated no significant evidence of publication bias (intercept, 0$40, 2 tailed; 95% CI, 22$69 to 3$48; P 5 0$778). The results of Duval and Tweedie trim-and-fill analysis suggested that the 2 studies were missing in the right side of the mean effect, and the adjusted OR obtained using the trim-and-fill approach was 0$92 (95% CI, 0$71-1$18). The result of classic fail-safe N suggested that 86 nil or null reports would be required to increase the P-value associated with the average effect to higher than an alpha level of 0$05. In addition, statin use was also not associated with reduced delirium-free days (pooled standard mean difference, 0$11; 95% CI, 20$41 to 0$93; P 5 0$450; see Figure S3, Supplemental Digital Content, http://links.lww.com/AJT/ A137). Funnel plot of studies on delirium-free days between statin users and nonusers was also shown in Supplemental Digital Content (see Figure S4, http:// links.lww.com/AJT/A137).
Using the GRADE approach, we provided the evidence of this meta-analysis a rating of low quality and a weak recommendation for this review.

DISCUSSION
In this meta-analysis, we reviewed the association between statin use and the risk of delirium among critically ill and surgical patients, and we did not observe a benefit of statin use for preventing delirium. The results are similar to those of the meta-analysis performed by Vallabhajosyula et al. 17 The overall quality of the evidence, as defined using the GRADE approach, was determined to be low with a weak recommendation, largely because of heterogeneity. Inflammatory alteration of the central nervous system is a major factor involved in the pathophysiology of delirium. 37,38 Statins may alleviate delirium through the attenuation of neuroinflammation, deleterious neurotransmitters, cerebral hypoperfusion, and microthrombosis, indicating the potential clinical application  of statins for the treatment of delirium. However, the clinical effects of statins for delirium remain a matter of debate. 39 Among critically ill patients, the use of statins in the ICU was associated with a decreased risk of delirium, especially in patients with early-stage sepsis; moreover, discontinuation of a previously used statin was associated with the increased occurrence of delirium. The anti-inflammatory effects of statins may confer neuroprotection, aid memory recovery, reduce the severity of cognitive dysfunction, and attenuate blood-brain barrier injury in critically ill and surgical patients. These pleiotropic effects may contribute to the prevention or mitigation of delirium in critically ill and surgical patients by modifying the process of neuroinflammation and the activation of proinflammatory microglia. [40][41][42] The results of the present subgroup analysis focusing on studies published before 2015 indicated that statin use was not associated with the risk of delirium (pooled OR, 1$07; 95% CI, 0$86-1$33; P 5 0$543; see   Figure S6A, Supplemental Digital Content, http:// links.lww.com/AJT/A137). This result is the same as that of a previous meta-analysis, 17 suggesting that statins may not reduce the occurrence of delirium in critically ill and cardiac surgery patients. However, the aforementioned meta-analysis included and reviewed only 6 observational studies published before 2015. The strength of our study is that 7 more studies, including 2 RCTs, were included than were in the previous meta-analysis. 17 In addition, our results indicated that statin use was associated with a decreased risk of delirium in 2 pooled RCTs, which were published after 2015 and conducted in Asia. We hypothesize that this finding would be associated with racial differences between Asians and other individuals. However, more studies are required to confirm this result and hypothesis. Basically, there should be less bias in randomized controlled trials than observational studies. We think all patients in these 2 randomized controlled trials having the same general anesthesia protocol and the same management could be another reason to avoiding bias and reveal the real effects of statin on delirium. The association of statin type and delirium needs more studies to confirm.
Our meta-analysis has several limitations. First, few studies were included in our analysis. In addition, heterogeneity among studies in the design and population may have biased the pooled results. Second, our study primarily analyzed observational studies, which have their own limitations. Detailed patients' baseline data, specifically an evaluation of organ dysfunction upon ICU admission and preadmission evaluation of delirium, were not uniformly reported in all studies. In addition, no confirmed information was provided on the timing, dose, duration, or type of statin therapy in these patients. Therefore, subgroup analyses stratified by the statin type could not be performed. Finally, the studies did not provide detailed information on the training of medical staff in the identification of delirium.
In conclusion, the present study conducted a systematic review and meta-analysis of studies published from 1975 to 2021, focusing on the role of statin therapy in reducing the risk of delirium among patients with critical illness or those undergoing surgery; the findings suggest a nonsignificant benefit of statin use in reducing the risk of delirium. Because of the multifactorial nature of delirium, many expected variations among studies were observed in patients, medical providers, and prescribed medications' characteristics. Further research-including the analyses of statins with different lipophilic or hydrophilic nature and half-lives, the timing of preoperative administration, and the timing of reinitiation-are required to clarify the association between statin use and delirium risk immediately after operation.

CONCLUSION
Statin use is not associated with the risk of delirium. Because of high heterogeneity, studies with better designs, especially RCTs and observational studies, are required to confirm the results of the present study. Statins are a potential therapy for delirium in critically ill patients, and considerable uncertainty remains in this topic. Box 1.