A recent paper published in JAMA Psychiatry provides important findings of the missing piece of clinical uncertainty about treatment and prevention in delirium.
Delirium, commonly known as acute confusional state, can often be caused by one or more contributing factors, such as a severe or chronic medical illness, changes in metabolic balance, adverse effects from medication, systemic infection, surgery, or alcohol or drug intoxication or withdrawal. It’s critically important to identify and treat delirium because some of the contributing factors could be life-threatening. However, there is no sufficient evidence for choice of medication to treat or prevent the symptoms of delirium.
A recent paper, Association of Delirium Response and Safety of Pharmacological Interventions for the Management and Prevention of Delirium A Network Meta-analysis, published in JAMA Psychiatry provides important findings of this missing piece in that important clinical uncertainty. The leading author of an international collaboration, Professor Kuan-Pin Su, at the China Medical University in Taichung, Taiwan, concludes the main finding about treatment/prevention of delirium: “In this report, we found that the combination of haloperidol and lorazepam demonstrated the best option for treatment of delirium, while ramelteon for prevention against delirium.” Hopefully, our findings would revise the chapter of treatments and prevention of delirium in the treatment guidelines. (https://jamanetwork.com/journals/jamapsychiatry/article-abstract/2726609).
In addition to the superiority of “haloperidol plus lorazepam” and “ramelteon” in treatment and prevention to delirium, the paper also revealed an important result that both of these regimens did not associate with higher all-cause mortality rate than placebo/control did. That is, the safety of such regimen was similar with that of placebo/control, in aspect of risk of mortality.
Delirious symptoms have profound negative impacts to patients and their caregivers, and significant increase the mortality and morbidity. According to another main author, Dr Ping-Tao Tseng from the WinShine Clinics in Kaohsiung, Taiwan, “it is equally important to treat underlying diseases (e.g. dehydration, adverse drug reaction, or metabolic imbalances) and to treat the neuropsychiatric symptoms (psychosis, irritable and unstable mood, agitated and destructive behaviors… etc.) of delirium. Therefore, the rationale of this work was to provide a general principle of medication prescription to manage the delirium symptoms rather than the etiology behind the delirium.” In another word, the results of current network meta-analysis were aiming to delirious symptoms but not targeting their underlying causes as the recruited subjects had different baseline condition among these 58 RCTs.
In addition to the superiority of “haloperidol plus lorazepam” and “ramelteon” in treatment and prevention to delirium, the paper also revealed the important findings about safety. Their result revealed that both of these regimens did not associate with higher all-cause mortality rate than placebo/control.
In the study by Wu and colleagues in a coming issue of JAMA Psychiatry, the authors use a variant of meta-analysis (network meta-analysis [NMA]) to explore pharmacologic interventions for delirium. Another main author, Dr Yi-Cheng Wu, from the Chang Gung Memorial Hospital in Linkou, Taiwan, added, “single randomized controlled trial or traditional pairwise meta-analysis provides less information than NMA to synthesize direct and indirect evidence for decision-making.”
The Editorial in the same issue by Professor Blazer from Duke University Medical Center, North Carolina, pointed out that this NMA is of importance for at least 2 reasons. First, non-pharmacological approaches for delirium are time-consuming and costly to implement. A safe and effective pharmacologic intervention would be welcomed. Second, new agents and new approaches to pharmacologic treatment have emerged that can be empirically investigated for efficacy and safety in the treatment of delirium (https://jamanetwork.com/journals/jamapsychiatry/fullarticle/2726605).
Indeed, non-pharmacological treatments are always important in the treatment and prevention in delirium. In fact, in this current NMA in JAMA Psychiatry, the authors have reported that the ramelteon, a melatonergic agonist, was ranked the best preventive regimen against delirium prevention. The authors added, “the action of melatonin agonists like ramelteon is to re-establish the disturbed circadian rhythm. Interestingly, the combination of simple behavioral interventions (e.g. exposure to diurnal light-dark cycles, familiar objects, clocks/calendars, …etc.) has been commonly advised to help to prevent delirium in clinical settings.
譫妄 (delirium)在重症治療中是一個相當重要的症候群，主要症狀包括突然其來嚴重的聽幻覺、視幻覺、及怪異妄想，合併意識混亂，讓照顧者以為患者中邪了。最新研究的報告指出（Pérez-Ros et al. Diseases 2019 Jan 30），譫妄在住院中老人盛行率高達27%，在其他重症病人中也非常常見（加護病房高達 35%, 如果使用上呼吸器更上達 80%；一般低風險手術後12%-13%；高風險手術後更高，例船冠狀心血管手術29%、重大腹部手術50%、心臟手術 51%；臨終病人高達85%！）由於譫妄患者死亡率高達20-30% (Lipowski NEJM 1989)，其臨床意義非常重要，代表個案的病情相當嚴重，不容輕忽！
目前醫學界公認對於治療譫妄最重要的就是找出潛在生理病因並盡快矯正（例如藥物中毒或戒斷、代謝系統異常、急慢性內外科疾病或感染病…等等）。然而，譫妄症狀（例如幻覺及妄想所造成的對自我及他人之危險以及治療和照顧上的困難），嚴重影響病人的預後，雖然其病因仍未完全明朗，研究認為譫妄幻覺妄想症狀與腦內多巴胺、乙醯膽鹼、GABA等神經傳導物質異常、以及生物節律與醒睡中樞異常有關。遺憾的是，目前臨床研究證據相當有限，療效不足、副作用、都增加臨床治療困難；此外，譫妄的預防更是讓臨床醫師爭論不休的議題。臨床治療指引 (clinical guideline) 都強調：目前的臨床研究證據不足，無法有足夠確實的結論！
中國醫藥大學精神醫學教授蘇冠賓醫師、林口長庚復健科吳易澄醫師、與高雄文信診所曾秉濤醫師等人主導，結合日本國家癌症中心、英國倫敦國王學院及加拿大多倫多大學之國際知名團隊，把現存所有譫妄藥物治療納入研究主題，利用網路統合分析（NMA）的最新方法，發現：褪黑激素受體藥物ramelteon對於預防重病患者產生譫妄症有顯著的功效！在由於褪黑激素受體藥物可以矯正生物節率和晝夜週期（circadian rhythm），其效果類似其他矯正譫妄症的非藥物療法，例如光照療法、利用時鐘或日曆強化人時地定向感…等，因此藥物如果可以有效地達到類似的效果，應該具有相當重要的臨床價值。生物的晝夜週期（circadian rhythm）的正常運作，不僅是身心健康的重要指標、是大腦記憶功能的守門員、更是維持代謝、免疫、心血管、骨骼肌肉功能的重要生理機能。此外，蘇教授團隊更首度證實Haloperidol與lorazepam兩藥物的組合最能改善譫妄症的症狀，這份重大突破的研究成果也獎刊登在國際知名期刊JAMA Psychiatry！這樣的結果，甚至獲得國外審查專家的評論：「做為未來臨床指引制定之重要根據！」
1. Lipowski, Z.J. Delirium in the elderly patient. N. Engl. J. Med. 1989, 320, 578–582.
[The outcome for the majority of patients is full recovery, but in 20 to 30 percent, delirium is followed by death], [About 40 percent of patients with dementia who were 55 years of age or older were delirious on admission to the hospital, whereas 25 percent of those who were delirious had dementia], [In older long-term care residents, the reported prevalence varies between 15% and 70%]
2. Pérez-Ros P, Martínez-Arnau FM. Delirium Assessment in Older People in Emergency Departments. A Literature Review. Diseases. 2019 Jan 30;7(1). pii: E14. doi: 10.3390/diseases7010014.
[The prevalence of delirium in end-of-life patients approaches 85%], [In hospitalized older patients, delirium is the most frequent complication, with an approximate prevalence of 27%], [In postsurgical patients, the general surgery pose lesser risk, with a prevalence of 12%-13%, whereas the patients subjected to aortic, major abdominal, or cardiac surgery is much higher, up to 29%, 50%, and 51%, respectively], [In the ICU, the prevalence varies between 31% and 35%, being higher in patients subjected to mechanical ventilation and with comorbidities (up to 80%)], [The probability of delirium increasing by 45% in moderate dementias and 58% in severe dementias].
REPLY to valuable comments about the limitation of NMA:
I am also very delighted to receive many valuable comments from excellent clinicians like you about our NMA in JAMA Psychiatry. They correctly pointed out the limitation of NMA and the caution of using FGS (haloperidol/lorazepem). I actually agree them all as well as the Editorial by Blazer in the same issue.
Although our study is strengthened by comparing different drugs with NMA, the generalization of our results are still highly limited and depending on the studies included in the NMA and the possible comparisons given the studies available. For example, there were many studies using haloperidol but only only one study compared haloperidol/lorazapam versus haloperidol/placebo. The result showing haloperidol/lorazapam the best because this combination was doing very well in Richmond Agitation-Sedation Scale (RASS) scores as compared to haloperidol/placebo in that single study.
In conclusion, we should always consider specific treatments in specific clinical conditions rather than applying only one general principle from a big picture of our NMA.
譫妄症狀 (delirium) 是什麼？
患者因為內外科問題而導致身體狀況惡化時，有一定機率會出現譫妄症狀的併發症，這時患者會出現急性且起起伏伏的躁動 (agitation) 、幻覺 (hallucination) 、妄想 (delusion) 、日夜混亂 (circadian disturbance)、胡言亂語 (irrelevant and incoherent speech) 等，而這些症狀出現時，往往代表患者的身體出現重大狀況！
在我們這篇最新的研究中，我們發現到Haloperidol與lorazepam的組合，是所有藥物處方中最能緩解譫妄症狀的處方，另一方面，如果不考慮複方，Haloperidol則是所有單方 (single regimen) 裡面效果最好的處方！而且根據統計，這些處方和安慰劑/對照組 (placebo/control) 比較起來，並不會因此造成更高的患者最終死亡率 (all-cause mortality rate)。
另一方面，我們也嘗試討論是否有任何處方能夠預防 (或減少) 譫妄症的發生，我們的研究發現ramelteon 這個藥物對於預防譫妄症狀的發生，是所有處方之中效果最好的，而Olanzapine則是所有抗精神病劑 (antipsychotics) 中預防效果最好的，而且這兩種藥物和安慰劑/對照組 (placebo/control) 比較起來，並不會因此造成更高的患者最終死亡率 (all-cause mortality rate)。 (圖片資料整理自原始文章)
至於其餘各種藥物 (例如大家常用的quetiapine) 的效果，以及各種藥物的使用劑量與時間等細節，就請有興趣的讀者自行上雜誌網站買全文來看了！ 最後，也提醒民眾，若真有遇到譫妄症狀的問題，切勿自行使用藥物，而應配合醫師接受積極檢查與治療方為上策唷！