Retour

Article pour les cliniciens

---

L'efficacité comparative des interventions pour les comportements agressifs et agités dans la démence: une revue systématique et méta-analyse en réseau.

---

---

• Watt JA
• Goodarzi Z
• Veroniki AA
• Nincic V
• Khan PA
• Ghassemi M, et al.

Lire le résumé Lire le texte intégral

---

---

---

Commentaires cliniques (en anglais)

General Internal Medicine-Primary Care(US)

Article is most relevant to hospitalists and SNFists; less relevant to primary care physicians who don`t provide care in these settings. A network diagram is used that is not a familiar tool nor really explained.

General Internal Medicine-Primary Care(US)

Interesting interventions for aggressive and agitated behavior in persons with dementia. It mirrors my anecdotal and clinical experience and should inform a "less is more" approach when it comes to using psychoactive drugs in folks who suffer from dementia.

Geriatrics

These results confirm that non-pharmacologic approaches should become standard of practice.

Geriatrics

This is a complex and ambitious network meta-analysis (NMA) of 163 studies comparing the efficacy of non-pharmacologic and pharmacologic interventions for physical and verbal aggression and physical and verbal agitation in 23,143 patients with dementia. The sample size was large enough to compare the efficacy of a variety of non-pharmacologic and pharmacologic interventions, using consensus-developed criteria for minimal improvement. The results (summarized in Figure 3) suggest that antipsychotics, which in the US have a black box warning for increasing the risk for mortality of older dementia patients and are associated with falls, sedation, and extra-pyramidal side-effects, should not be first-line therapy for dementia-related agitation and aggression. The one area in which pharmacotherapy (anticonvulsants) proved superior was in the realm of verbal agitation. Of note, typical antipsychotics showed overall low benefits, and atypical antipsychotics emerged as poorly effective except for verbal agitation, for which they were less effective than anticonvulsants. The take-home message is that there is little role for antipsychotics in the management of agitation and aggression despite the high prevalence of use for these indications. The studies, however, were at high risk for bias due to missing data on the dominant outcome variable, subsets of the Cohen-Mansfield Agitation Inventory, as well as representing mostly women in nursing homes = age 80. This bias substantially decreases the generalizability of the results, but they nonetheless underscore that non-pharmacologic approaches that involve physical activities (outdoor activities in particular) and massage and touch therapy may be relatively the most successful. Because it is a meta-analysis, the study does not attempt to evaluate efficacy by whether agitation (and also aggression) appear driven by hallucinations, delusions, or anxiety/depression. Another limitation is that the study does not include antidepressants among the pharmacologic agents studied despite the positive effect of citalopram in reducing the severity of agitation in dementia patients.