荷科学家开发出ICU患者谵妄预测模型

荷兰内梅亨大学医学中心的Mark van den Boogaard博士及其同事报告称，一种名为PRE-DELRIC(ICU患者谵妄预测)的谵妄预测模型在成人重症监护病房(ICU)患者中的预测效能明显优于医生和护士预测方法，将有助于ICU患者谵妄的预防(BMJ 2012;344:e420 [doi: 10.1136/bmj.e420])。

研究者称，谵妄预防措施可控制其发生率、严重程度以及持续时间。虽然有数种评估工具已用于预测其他科室住院患者的谵妄风险，但目前尚没有一种可用于ICU患者的循证模型。对所有ICU患者采取常规预防措施不仅费时，而且有可能使许多患者面临药物预防相关不良事件等不必要的风险。PRE-DELRIC模型可在患者入住ICU后24 h内对10种风险因素进行评估：年龄、APACHE II(急性生理和慢性健康状况评分II)评分、入住类别、昏迷、感染、代谢性酸中毒、吗啡使用、镇静剂使用、尿素浓度以及紧急入住情况。通过该模型可确认高危患者并对其实施预防性治疗，这将非常有助于重症监护实践，并确保高危患者有效利用研究资源。在临床方面，该模型可促进高危患者改善定向、认知刺激、早期活动以及听音乐等非药物预防措施的应用。

对于非危重症患者，非药物预防措施可减少谵妄发生率和持续时间，氟哌啶醇治疗可降低严重程度、持续时间和与之相关的住院时间。但对于ICU患者而言，缺乏相关数据，而PRE-DELERIC模型可为这些患者提供氟哌啶醇预防用药的选择信息。既往研究(Lancet 2009;373:1874-82)表明， ICU机械通气患者早期活动除产生其他方面的显著效果外，还可减少谵妄持续时间。

研究者在对该模型进行临时验证后，利用2009年1月1日~9月1日入住荷兰4家医院ICU患者数据进行外部验证研究，合并数据包括3,056例≥18岁患者信息。结果显示，操作曲线下面积(AUROC)为0.85。依据风险大小，将患者分为低、中、高和非常高风险4组，其敏感性和特异性分别为81% 和 75%、62% 和 89%、46% 和 95%、30% 和 98%。

研究者还利用124例便利样本，比较了该模型与医护人员对患者谵妄预测结果。结果显示，前者AUROC(0.87)显著高后者(0.59)。此外，ICU护士与实习护士、重症专科医生、资深重症专科医生、住院医生的预测结果无显著差异。

研发PRE-DELIRIC模型的医院正在日常实践中应用该模型，并对谵妄高危ICU患者(PRE-DELIRIC>50%)以及痴呆或酒精滥用患者采取预防措施。但ICU患者最佳临界点和最有效干预措施尚有待进一步研究。此外，上述研究的局限性在于该模型的静态特性，不能反映患者健康状况的变化，而后者可影响谵妄风险。

研究者报告无相关利益冲突。

爱思唯尔  版权所有

A recently developed tool could help doctors stay ahead of the game in preventing delirium in intensive care patients.

Dutch researchers say their delirium prediction model, known as PRE-DELRIC, was significantly more successful than doctors and nurses at predicting delirium in hospitalized adults.

Preventive measures for delirium can limit its incidence, severity, and duration. While several assessment tools exist for other segments of hospitalized patients, "no evidence-based prediction model for general intensive care patients is available," Mark van den Boogaard, Ph.D., of Radboud University Nijmegen (Netherlands) Medical Centre and his colleagues said (BMJ 2012;344:e420 [doi: 10.1136/bmj.e420]).

General preventive measures in all ICU patients are time consuming, and may expose many patients to unnecessary risks such as adverse events related to drug prophylaxis, the researchers explained.

For PRE-DELRIC (Prediction of Delirium in ICU Patients), Dr. van den Boogaard and his colleagues defined 10 risk factors that can be easily assessed within 24 hours of admission to the ICU: age, APACHE II (Acute Physiology and Chronic Health Evaluation II) score, admission category, coma, infection, metabolic acidosis, morphine use, sedative use, urea concentration, and urgent admission.

"The use of the PRE-DELERIC model to identify and consequently preventively treat high-risk patients could offer an important contribution to intensive care practice and ensure efficient use of research resources to study only high-risk patients," the researchers said.

Clinically, the model may improve the use of nondrug measures to prevent delirium in high-risk patients, the researchers added. Such measures include improvement of orientation, cognitive stimulation, early mobilization, and listening to music, they said.

In noncritical patients, nondrug preventive measures have been shown to reduce delirium incidence and duration, and haloperidol treatment has lessened severity, duration, and associated length of stay. But for ICU patients, data are hard to come by. PRE-DELERIC could inform the choice to use prophylactic haloperidol in these patients, the authors said. Existing research (Lancet 2009;373:1874-82) does show that "early mobilisation of mechanically ventilated patients in intensive care, besides other significant effects, resulted in a reduced duration of delirium," Dr. van den Boogaard and his coauthors wrote.

After testing their model for temporal validation, the researchers conducted an external validation study of data from intensive care patients admitted to four Dutch hospitals between Jan. 1 and Sept. 1, 2009. The pooled data included information from 3,056 patients aged 18 years and older, yielding an area under the receiver operating characteristics curve (AUROC) of 0.85. The patients were divided into four risk groups: low, moderate, high, and very high. The sensitivity and specificity were, respectively, 81% and 75% for the group with low-risk group; 62% and 89% for the moderate-risk group; 46% and 95% for the high-risk group; and 30% and 98% for the group with very high risk.

The researchers compared the predictions of patient delirium made by their model to predictions made by doctors and nurses in the hospital, using a convenience sample of 124 patients.

The AUROC for the PRE-DELIRIC model (0.87) was significantly higher than that of the doctors and nurses (0.59).

No significant differences appeared in the predictions of intensive care nurses compared with student intensive care nurses or among intensivists, fellow-intensivists, and residents, the researchers said.

The PRE-DELIRIC model is being used in daily practice in the hospital where the model was developed, the researchers said. "Intensive care patients with a high risk of delirium (at least a 50% PRE-DELIRIC score), and patients with dementia or alcohol misuse, receive preventive measures. The optimal cut-off point of the PRE-DELIRIC model and the most effective delirium preventive interventions for intensive care patients need to be studied in the near future."

The findings were limited by the static nature of the model, which does not account for changes in health status that might affect the odds of developing delirium, the researchers noted.

The researchers reported having no financial conflicts of interest.