Pain, agitation, and delirium practices in Chinese intensive care units: A national multicenter survey study

To the Editor: The standardized management of pain–agitation–delirium (PAD) represents a fundamental component of clinical practice in intensive care units (ICUs).[1] Evidence-based analgesia and sedation strategies not only alleviate patients’ physiological discomfort and psychological distress, but also significantly reduce mechanical ventilation duration, improve long-term prognosis, and optimize healthcare resource allocation.[1,2] Equally crucial is delirium management, as evidenced by studies demonstrating that delirium duration independently predicts mortality risk, hospital stay length, healthcare expenditures, and acquired dementia incidence.[3] Although existing studies have explored PAD management practices, most rely on clinicians’ subjective recall that may distort objective clinical realities and fail to reflect distinctive characteristics.[4] Therefore, this nationwide multicenter cross-sectional study, initiated by the Critical Care Analgesia and Sedation Quality Control Subspecialty Group under the China National Center for Medical Quality Control in Critical Care Medicine, aims to investigate the current status of PAD management in China. The findings would establish an evidence-based foundation for developing national quality control indicator systems tailored to China’s clinical realities. This nationwide multicenter cross-sectional survey (No. ChiCTR2400093986, https://www.chictr.org.cn), approved by the West China Hospital Biomedical Ethics Committee (No. 2024-1666), evaluated PAD management practices across China on August 14, 2024. Using stratified sampling by hospital grade and type, seven hospitals per province were selected, enrolling all eligible ICU patients during the 24-hour study period [Supplementary Figure 1, https://links.lww.com/CM9/C607]. Data collection conducted at 10:00 and 20:00, with reduced evening participation reflecting post-morning ICU transfers. All the clinicians participated voluntarily and anonymously, with survey data anonymized and used solely for research purposes, ensuring no disclosure of personal information (patient identifiers were anonymized through dual-encryption methodology, and database access was tier-restricted through role-based permission protocols, with audit trails documenting all queries). Retrospective analysis of anonymized data followed the Declaration of Helsinki principles, with waived informed consent and voluntary clinician participation. Provincial distributions and case allocations are detailed in Supplementary Tables 1 and 2, https://links.lww.com/CM9/C607. This cross-sectional study collected data from 166 hospitals across 31 provincial-level administrative regions in China. The same patient cohort underwent dual assessments at 10:00 (n = 2627) and 20:00 (n = 2213), with baseline demographic and clinical characteristics at 10:00 detailed in Supplementary Table 3, https://links.lww.com/CM9/C607. At 10:00, the assessment implementation rates for analgesia, sedation, and delirium were 81.61% (n = 2144), 60.03% (n = 1577), and 72.40% (n = 1902), respectively [Supplementary Table 4, https://links.lww.com/CM9/C607]. There were no significant differences in the sedation assessment rates among the three hospital grades (P = 0.64), while delirium assessment rates showed significant variations (P <0.01). For the analgesia assessment, the observed difference approached statistical significance (P = 0.05). Therefore, we suggest that this finding may indicate potential variations in clinical practice, though further validation with larger sample size is needed (the predefined significance threshold for this study was α = 0.05). With respect to the utilization of PAD assessment tools, at 10:00, analgesia assessment tools were used 2226 times in total, with Critical-Care Pain Observation Tool (CPOT) being the predominant tool, accounting for 1380 (61.99%). Sedation assessment tools were used 1667 times, with the Richmond Agitation-Sedation Scale (RASS) being the most widely adopted tool, used 1399 times (83.92%). Delirium assessment tools were applied 2089 times, with the CAM-ICU being the most frequently used tool (1030 instances, 49.31%), followed by subjective judgment, which accounted for 42.51% (n = 888) of the assessments. The results collected at 20:00 were similar to those obtained at 10:00 [Supplementary Table 5 and Supplementary Figure 2, https://links.lww.com/CM9/C607]. Midazolam (n = 178/716, 24.86%) and dexmedetomidine (n = 155/655, 23.66%) were the sedations most commonly used monotherapies at 10:00 and 20:00, respectively. Sedative medications were frequently administered in combination, with 17.87% (n = 128/716) of patients at 10:00 and 17.71% (n = 116/655) at 20:00 receiving multiple sedations. The most prevalent combination of sedatives was propofol and midazolam, which were used by 6.84% (n = 49/716) and 7.18% (n = 47/655) of patients at 10:00 and 20:00, respectively. For analgesia, remifentanil was the most commonly used monotherapy, with 51.47% (n = 490/952) and 51.90% (n = 423/815) of patients receiving it at 10:00 and 20:00, respectively. Multimodal analgesia was rarely used [Supplementary Figure 3, https://links.lww.com/CM9/C607]. In the study, sedative medications were administered a total of 1615 times, whereas analgesic medications were administered 1789 times [Supplementary Figure 4, https://links.lww.com/CM9/C607]. The most common unit of administration for both sedative and analgesic medications was “mL/h”, accounting for 64.95% (1049/1615) and 60.08% (1075/1789) of usage, respectively. The second most commonly used dosage unit for sedation was “mg/h” (218/1615, 13.50%), whereas for analgesia, it was “µg/h” (272/1789, 15.20%). However, the internationally standardized dosage unit for both sedatives and analgesics is “µg·kg–1·h–1”, which was used in this study at rates of 7.00% (n = 113/1615) for sedation and 8.72% (n = 156/1789) for analgesia. The results of this survey revealed that the most commonly used pain and sedation assessment scales are the CPOT and RASS. This finding is consistent with the 2018 PADIS guidelines (Pain, Agitation/Sedation, Delirium, Immobility, and Sleep Disruption). The CPOT is the most robust scale for assessing pain in critically ill adults unable to self-report, and it has very good psychometric properties with the highest scores.[1] The RASS is the first sedation scale validated to detect changes in sedation status over consecutive ICU days.[5] This study revealed that the usage rates of propofol and midazolam are the highest during the day, whereas dexmedetomidine is most frequently used at night. This result slightly differs from the recommendations in the guidelines. The guidelines recommend using either propofol or dexmedetomidine instead of benzodiazepines, as this can shorten the duration of mild sedation by at least 4 hours and the extubation time by 8–12 hours.[1] However, in our study, the usage rate of midazolam was relatively high during the day, which is inconsistent with the guidelines. This may suggest that we are not optimizing the selection of sedatives as recommended, and the reason we hypothesize that this preference may stem from midazolam’s pharmacological profile combined with the allocation of medical resources, particularly in primary care settings or resource-constrained clinical environments with limited monitoring capabilities. However, the finding that dexmedetomidine was most frequently used at night aligns with the guideline recommendations.[1] A large body of research shows that the use of dexmedetomidine can reduce the incidence of delirium, and compared with the use of midazolam, the use of dexmedetomidine is associated with shorter durations of mechanical ventilation.[6] This study revealed that when the two time points (10:00 and 20:00) were combined, the number of sedative drug administrations (1615 times) was lower than the number of analgesic drug administrations (1789 times). The 2018 PADIS guidelines recommend prioritizing analgesia (i.e., using analgesic drugs, typically opioids, before sedatives to achieve the desired level of sedation) or using analgesia-based sedatives (i.e., using analgesic drugs, usually opioids, in place of sedatives to achieve sedation).[1] This study used opioids such as remifentanil, sufentanil, and fentanyl, which are in line with the guidelines that emphasize the role of opioids in pain management for ICU patients. The most frequently used analgesic in this study was remifentanil, which may be due to its rapid onset, short duration of action, mild respiratory depression, and metabolism that is less influenced by liver or kidney function. The study on analgesia and sedation has used the unit “µg·kg–1·h–1” for drug dosing.[4] However, the PADIS guidelines do not explicitly recommend “µg·kg–1·h–1” as the standard unit. Currently, the implementation of this dosing unit in analgesia and sedation clinical practice in China is also suboptimal. This may be because many clinicians adopt an outcome-oriented approach to analgesia and sedation, with insufficient attention and awareness regarding the precise dosing of these medications. In addition, these practice patterns are also inextricably linked to limited exploration of inter-hospital variability in critical determinants such as clinician training levels and resource allocation disparities. Given the potential risks of under- or overdosing due to inconsistent dosage units, standardized training programs for physicians and nurses should be implemented regularly. In addition, the adoption of electronic medical record alert systems could be considered to enhance the precision of analgesic and sedative dose management. An important aspect of this study is related to delirium. This study suggests that ICU healthcare providers in China may not be sufficiently attentive to delirium. The data from 10:00 show that 27.60% (725/2627) of patients had not been assessed for delirium, and the number of patients who had not been assessed for delirium at 20:00 was even greater, at 30.14% (667/2213) [Supplementary Tables 3 and 4, https://links.lww.com/CM9/C607]. In the adult critical care population, delirium represents a frequently encountered clinical phenomenon that contributes to considerable patient and family distress while being independently associated with poorer clinical outcomes.[3] We propose that this phenomenon may be attributed to multiple systemic factors, including healthcare resource constraints (excessive clinician workload and limited specialized monitoring equipment availability) and educational gaps (insufficient delirium assessment protocol training and under-recognition of delirium’s prognostic significance). This study, while comprehensive, has several limitations. First, the cross-sectional design restricts causal inference, limiting the analysis of the relationships between PAD management and patient outcomes. Future studies should incorporate longitudinal data. Second, sample sizes in some regions may be small, potentially affecting the representativeness of the data, particularly in economically underdeveloped areas. Third, self-reported data may introduce reporting bias, and there are variations in implementation standards across hospitals. Future research could validate these findings through direct observation or patient records. Finally, not all potential confounders, such as patient disease type and comorbidities, were fully controlled, which may have influenced treatment choices and outcomes. Future studies should address these variables. The study results indicate that, overall, Chinese practices in PAD management are largely consistent with international guidelines. However, certain aspects, particularly in the finer details of implementation, still require ongoing refinement and improvement to fully align with global standards. Therefore, regular training and academic exchanges should be prioritized to enhance team training and professional capabilities. In addition, leveraging research to support clinical protocols and development is crucial. Most importantly, stringent quality control should be emphasized to ensure continuous progress. Members of group We thank the members of the Analgesia and Sedation Quality Control Subspecialty Group, National Center for Medical Quality Control in Critical Care Medicine, for their contributions. The complete list is provided in the supplementary files, https://links.lww.com/CM9/C607. Funding This work was supported by a grant from 1.3.5 project for disciplines of excellence, West China Hospital, Sichuan University (No. ZYGD23012). Conflicts of interest None.