, Introduction
Delirium, an acute and preventable confusional state, poses significant challenges in the postoperative care of geriatric patients undergoing spinal surgery. With its occurrence in up to 50% of older adults postoperatively [1–5], delirium represents a critical issue associated with adverse outcomes such as prolonged hospital stays, increased healthcare resource utilization, cognitive decline, and heightened mortality rates [4–7]. Despite efforts to manage delirium reactively, its duration, severity, and long-term consequences remain largely unaffected by such strategies, underscoring the importance of preventive approaches.
Given the increasing recognition of delirium as a serious postoperative complication, numerous multicomponent nonpharmacologic interventions have been proposed to reduce its incidence, including the regulation of sleep-wake cycles, continuous use of sensory aids, frequent re-orientation, early mobilization, increased family involvement, and fluid management [8]. However, the effectiveness of these interventions in geriatric spinal surgery patients remains uncertain, necessitating further investigation including risk factor-based intervention.
With increasing interest in postoperative delirium following spine surgery in the United States, Japan, and elsewhere, research on this topic has become more active. For instance, Pernik et al. [9] investigated risk factors for postoperative delirium, identifying age, gender, duration of surgery, and intraoperative blood loss as significant factors. This heightened interest is evident in studies conducted in various countries. Shi et al. [10] conducted a meta-analysis and identified common risk factors based on published papers from multiple countries.
However, relative to papers from Japan and China, Korean literature on this topic appears relatively sparse, with Korean studies primarily conducted in single institutions. Additionally, the lack of consistency in diagnostic methods has posed challenges in identifying diverse risk factors [11].
Considering this situation, our study aims to explore the prevalence of delirium and risk factors associated with delirium incidence in geriatric spinal surgery patients through a multicenter retrospective observational study. Specifically, we seek to observe variations in delirium rates across institutions and assess potential differences in operational definitions of delirium. Furthermore, we aim to provide insights into the importance of preoperative patient and caregiver education and management, particularly in light of recent emphasis on these aspects following geriatric spinal surgery.
Material and Method
Inclusion of patients
This retrospective analysis included data from 7 independent institutions between January 2022 and September 2023. At A, D, and E centers, 2,157 patients underwent spinal fusion for degenerative lumbar disease, of whom 347 were aged 75 years or older (A, 78/440; D, 110/956; E, 159/761). For the remaining institutions (B, C, F, and G), the total number of surgeries could not be disclosed due to institutional policies, but the number of patients aged 75 years or older was available: B, 511; C, 187; F, 42; and G, 201. In total, 943 patients aged 75 years or older were identified across the 7 centers. After excluding cases with missing variables, 1,284 individuals were included in the final analysis out of 1,290.
Diagnostic criteria for delirium and analyzed risk factors
The prevalence of delirium was identified based on the Diagnostic and Statistical Manual of Mental Disorders-5 (DSM-5) diagnostic criteria (Table 1) [12], where physicians recorded the diagnosis in the electronic medical records, prescribed antipsychotic medications, or conducted consultations with the Department of Psychiatry.
Preoperative and intraoperative risk factors for postoperative delirium were evaluated, including demographics (age, sex, height, weight), lifestyle (smoking), comorbidities (hypertension, diabetes, heart failure), and medication history (opioid and anticoagulant use). Those factors were compared in patients with and without delirium.
Statistical analysis
Statistical differences between 2 groups were assessed using the Student’s t-test or the Mann-Whitney U-test for continuous variables, and the chi-square test for categorical variables. All statistical analyses were performed using the latest version of SPSS software for Windows ver. 22.0 (IBM Corp.). A significance level of P<0.05 was deemed significant for all analyses.
Results
The study’s results across the 7 centers, designated A through G, focused on the presence or absence of a delirium diagnosis, the administration of the medications quetiapine fumarate and haloperidol, and the provision of psychiatric consultations. The data were divided into 2 subgroups: those with a delirium diagnosis or receiving the intervention (Yes) and those without a diagnosis or not receiving the intervention (No).
Among the 1,284 enrolled patients, 154 were identified as having delirium according to the operational diagnostic criteria. For the subgroup with a delirium diagnosis (labeled ‘Yes’ in DSM-5 diagnosis), Center D showed the highest proportion with 36 cases (33.3%), reflecting a significant occurrence of delirium diagnoses compared to other centers. In contrast, the lowest occurrences were in Center E with 6 cases (3.8%) and Center C with 10 cases (5.3%). Overall, the percentage of diagnosed cases was lower than those without a diagnosis, indicating that the majority of the population across centers did not meet the coincidence at postoperative general psychiatric condition.
When examining the use of medication (medication), the subgroup ‘Yes’ indicates those who were administered quetiapine fumarate or haloperidol. In the medication subgroup analysis, inter-center variability became apparent when antipsychotic administration was examined relative to DSM-5–diagnosed delirium cases. Centers A, C, and F demonstrated complete concordance, with all DSM-5–diagnosed patients receiving antipsychotic treatment (100%). In Centers D and G, nearly all diagnosed patients were treated (97.2% and 96.2%, respectively). Interestingly, Centers B and E prescribed antipsychotics to more patients than were identified by DSM-5 criteria (108.3% and 283.3%, respectively), indicating that treatment was extended to suspected cases even in the absence of a documented DSM-5 diagnosis. These findings highlight marked inter-center variability in the reliance on pharmacologic management of delirium.
In the context of psychiatric consultations (psychiatric consult), subgroup ‘Yes’ represents those who received consultations. By contrast, in the psychiatric consultation subgroup analysis, more marked variation was observed across centers when consultations were examined relative to DSM-5–diagnosed delirium cases. Centers A, C, and F demonstrated complete concordance, with all DSM-5–diagnosed patients receiving psychiatric consultation (100%). In Center D, less than half of the diagnosed patients were referred for consultation (44.4%). In contrast, Centers B and E recorded more consultations than DSM-5–diagnosed cases (144.4% and 316.7%, respectively), indicating that psychiatric involvement was extended to patients without a documented DSM-5 diagnosis. Center G showed an extremely low rate, with only 1 of 52 DSM-5–diagnosed patients (1.9%) receiving psychiatric consultation. These findings underscore substantial inter-center variability in the role and utilization of psychiatric services in delirium management (Fig. 1).
Table 2 provides a detailed overview of the demographic and clinical risk factors of the postoperative geriatric delirium, divided into groups without delirium (non-delirium) and with delirium (delirium). Statistically significant differences between the delirium and non-delirium groups were observed for age (78.7±3.9 vs. 79.5±3.5 years, P=0.012), sex (male, 470 vs. 83; P=0.005), height (157.6±9.9 vs. 159.8±9.5 cm; P=0.010), and smoking status (69 vs. 21, P=0.001). In contrast, no significant differences were found for weight, hypertension, diabetes, heart failure, anticoagulant use, or general endotracheal anesthesia (GEA) history. Opioid history demonstrated a very significant association with delirium, with a P<0.001. These findings highlight potential risk factors and correlations that may contribute to the development or identification of delirium in geriatric spine patients before surgery
In the risk factor analysis, age, sex, height, smoking history, and opioid use history were significantly associated with delirium. Older age (odds ratio [OR], 1.10; 95% confidence interval [CI], 1.02–1.19 per year), male sex (OR, 1.61; 95% CI, 1.12–2.31), taller height (OR, 1.05; 95% CI, 1.01–1.09 per cm), smoking history (OR, 2.45; 95% CI, 1.46–4.12), and opioid use history (OR, 2.54; 95% CI, 1.74–3.72) were identified as significant risk factors for postoperative delirium (Fig. 2).
To account for potential confounding among these variables, we additionally performed a multivariate logistic regression analysis. The results, summarized in Fig. 3, demonstrated that older age, male sex, and a history of opioid use remained significant independent predictors of postoperative delirium after adjustment. Specifically, older age was associated with an increased risk (adjusted OR, 1.08; 95% CI, 1.01–1.16; P=0.023), male sex conferred higher risk compared with female sex (adjusted OR, 1.42; 95% CI, 1.03–1.96; P=0.031), and opioid use history was the most potent predictor (adjusted OR, 2.18; 95% CI, 1.45–3.25; P<0.001). In contrast, height (adjusted OR, 1.01; 95% CI, 0.97–1.05; P=0.612) and smoking history (adjusted OR, 1.53; 95% CI, 0.95–2.45; P=0.079), which were significant in the univariate analysis, did not retain statistical significance in the multivariate model, indicating that their apparent effects were largely explained by interrelationships with other variables. Anticoagulant use likewise showed no independent association (P=0.667).
Weight, hypertension, diabetes, heart failure, anticoagulants, and GEA history did not show a statistically significant association with delirium presence, as their p-values were above 0.05. Opioid history demonstrated a very significant association with delirium, with a P<0.001. These findings highlight potential risk factors and correlations that may contribute to the development or identification of delirium in geriatric spine patients before surgery.
Discussion
We retrospectively observe the multifaceted nature of delirium in geriatric patients post spinal surgery, as illuminated by the study’s findings across 7 medical centers, labeled A through G. The study meticulously cataloged instances of delirium, the administration of antipsychotic medications such as quetiapine fumarate and haloperidol, and the engagement of psychiatric consultations, discerning a clear demarcation between patients diagnosed with delirium and those who were not, as well as between patients who received specific interventions versus those who did not.
The incidence rates of delirium observed in our study align with those reported in the literature, albeit with some variation. For instance, Kawaguchi et al. [13] reported a delirium incidence of 12.5% among spinal surgery patients, which is comparable to the overall incidence of 12.0% in our study. However, the incidence varied across centers, with the highest at 33.3% in Center D and much lower rates in Centers E and C, suggesting potential variability in diagnostic practices, patient demographics, or care protocols across institutions.
The considerable variability in delirium incidence across centers may be explained by several factors. First, although all cases were identified using DSM-5 diagnostic criteria, differences in how these criteria were applied and interpreted by physicians may have influenced the reported incidence, as a degree of subjectivity is inherent in the assessment process. Second, heterogeneity in patient populations, including comorbidities, nutritional status, baseline cognitive function, and the types of surgical procedures performed, may contribute to differential vulnerability to delirium. Third, variations in surgical and anesthetic protocols, such as operation time, blood transfusion, and perioperative use of opioids or sedatives, are known to affect delirium risk [4,9,10]. These factors together likely underlie the inter-institutional differences observed in our study.
The role of medications, particularly quetiapine fumarate and haloperidol, in managing or potentially contributing to delirium risk, is a critical area of investigation. Our findings demonstrate considerable variability in the use of these medications across centers. In some institutions, the proportion of patients receiving antipsychotics exceeded the number diagnosed with delirium according to DSM-5 criteria, whereas in others, medication use was relatively lower. This variation reflects differences in prescribing practices and clinical decision-making among centers. Nevertheless, medications were administered to the majority of patients diagnosed with delirium, indicating that antipsychotic treatment was commonly employed as part of delirium management in this cohort.
When calculated against the number of patients diagnosed with delirium, psychiatric consultations were performed in 72.7% of cases overall (112/154). In addition, the proportion varied markedly across centers, ranging from only 1.9% in Center G to over 300% in Center E. Compared with the nearly universal use of medications (165/154, 107.1%), the lower and inconsistent rate of psychiatric consultations (112/154, 72.7%) suggests that psychiatric involvement in delirium management may be underestimated across institutions. This finding underscores the importance of psychiatric oversight in postoperative care for older adults, consistent with Inouye et al. [14], who emphasized the effectiveness of multidisciplinary interventions in preventing and managing delirium. These discrepancies highlight the need to establish clearer criteria and standardized practices for psychiatric consultation.
Drawing upon the comprehensive demographic and clinical risk factor analysis presented in Fig. 2, it is evident that certain factors such as age, sex, height, smoking status, and opioid history significantly correlate with the incidence of delirium, underscoring the importance of a nuanced preoperative assessment to identify high-risk patients. Conversely, factors like weight, hypertension, diabetes, heart failure, anticoagulant use, and GEA history did not exhibit a statistically significant association with delirium, indicating that while these factors contribute to the overall clinical picture, they may not directly influence the likelihood of delirium in a postoperative setting.
Older age and taller height demonstrated modest effect sizes per 1-year and 1-cm increment, respectively; however, when considered over broader increments, the clinical impact becomes more apparent—for instance, a 10-year increase in age corresponds to an approximately 2.6-fold higher risk, and a 5-cm increase in height corresponds to an approximately 1.3-fold higher risk. Male sex, smoking, and particularly opioid use history showed stronger associations, with opioid use emerging as the most potent predictor of postoperative delirium (P<0.001). The critical role of opioid exposure in delirium development is consistent with the observations of Robinson et al. [4], who identified opioids as a major risk factor for postoperative delirium.
Nevertheless, it should be noted that these odds ratios were derived from univariate analyses. Potential confounding among the variables cannot be excluded, as sex, height, and smoking history are likely interrelated with male patients generally being taller and more likely to smoke. To address this issue, we additionally performed a multivariate logistic regression analysis, the results of which are summarized in Fig. 3. This analysis demonstrated that older age, male sex, and a history of opioid use remained significant independent predictors of postoperative delirium after adjustment, whereas height and smoking history lost statistical significance, suggesting that their apparent effects were largely explained by interrelationships with other variables. The strong association between opioid history and postoperative delirium in particular underscores the clinical importance of perioperative pain management strategies. Excessive reliance on opioids for analgesia may increase the risk of delirium, while undertreated pain itself can also contribute to its occurrence. Therefore, a balanced approach is essential. Opioid-sparing strategies—such as multimodal analgesia using non-opioid medications, the incorporation of regional anesthesia techniques, and perioperative protocols that promote early mobilization—should be considered to minimize opioid exposure while maintaining adequate pain control [15–17]. Implementation of such measures may help reduce delirium risk and improve outcomes in geriatric patients undergoing spinal surgery.
These findings are generally consistent with previous reports and provide further support for the existing body of evidence. First, with respect to opioid use, prior studies have demonstrated its strong association with postoperative delirium, which is corroborated by our results [4,18,19]. Second, regarding sex, several investigations have reported that male sex increases the risk of delirium, in line with the present analysis [20–22]. In contrast, although smoking history has been reported as a significant risk factor in other studies, our multivariate analysis revealed only a nonsignificant trend (P=0.079) [23,24]. This discrepancy may be attributable to the inherent limitations of a retrospective design and the lack of detailed quantitative assessment of smoking exposure such as current vs. former smoking status, duration, and intensity, which could have influenced the observed results.
Finally, this study underscores the complex interplay of factors contributing to postoperative delirium in geriatric patients undergoing spinal surgery. They advocate for a multidimensional approach to preoperative assessment, emphasizing the identification and mitigation of modifiable risk factors. Furthermore, the variability in delirium incidence, medication administration, and psychiatric consultations across centers calls for a collaborative effort to establish standardized guidelines and protocols that can enhance the quality of care and patient outcomes in this demographic.
This study has several limitations that should be acknowledged. First, its retrospective design may have introduced selection bias and limited the accuracy of delirium diagnosis. Because the data were collected retrospectively rather than through a standardized delirium research protocol, validated screening tools such as the Confusion Assessment Method could not be uniformly applied across centers. Second, important factors such as baseline cognitive function, nutritional status, and detailed quantification of smoking exposure, including current versus former status, duration, and intensity, were not available, which may have led to residual confounding. Third, information on preoperative use of gabapentinoids, which have been implicated in increasing delirium risk in older patients while also serving as opioid-sparing agents, was not collected; this dual role underscores the need for further investigation. Fourth, although pharmacologic interventions were discussed, non-pharmacologic preventive strategies and perioperative ethical aspects such as preoperative cognitive screening and patient safety protocols could not be assessed, and future prospective studies should address these elements. Finally, the study was limited to 7 centers, and differences in institutional practices, documentation systems, and the availability of psychiatric support may restrict the generalizability of our findings.
In conclusion, this study not only sheds light on the significant demographic predictors of postoperative delirium but also sets the stage for future research aimed at refining surgical and postoperative care practices for elderly patients undergoing spinal surgery. The current understanding of risk factors and the inter-center variability observed in this study provide a valuable foundation for developing targeted preoperative informed consent and preventive strategies to reduce the incidence and impact of delirium in this high-risk population.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
