**3.3 Symptoms in the late phase**

Following initial neurological deficits after acute CO intoxication, some patients experience progressive neurological deterioration, while others nearly complete recovery of symptoms. Some patients have a delayed onset of neurological deficits after an initial symptom-free period (Lee and Marsden 1994). The latter is often termed as delayed neuropsychiatric sequela in CO intoxication. The lucid interval after acute CO poisoning, on average, is around 20 days, varying from one to 240 days (Choi 1983; Lee and Marsden 1994; Ernst and Zibrak 1998; Pavese, Napolitano et al. 1999; Hsiao, Kuo et al. 2004), with a prevalence of 0.2- 40% (Hsiao, Kuo et al. 2004; Otubo, Shirakawa et al. 2007). Delayed neuropsychiatric sequelae include parkinsonism (Lee and Marsden 1994), chorea (Park and Choi 2004), akinetic mutism (Lee and Marsden 1994), increased irritability, verbal aggressiveness, violence, impulsiveness (Meredith and Vale 1988), mood disorders (Weaver 2009), dementia (Meredith and Vale 1988; Ernst and Zibrak 1998; Weaver 2009), psychosis (Ernst and Zibrak 1998), sleep disturbances (Weaver 2009), cortical blindness (Quattrocolo, Leotta et al. 1987; Senol, Yildiz et al. 2009) and incontinence (Ernst and Zibrak 1998).

The cognitive deficits are often very diverse (Hurley, Hopkins et al. 2001; Parkinson, Hopkins et al. 2002; Raub and Benignus 2002) including impairment in verbal or visual episodic memory, language, visuospatial ability, executive function and calculation (Chang, Chang et al. 2010). No specific neuropsychiatric battery has been designed for the cognitive deficits in CO intoxication. For general cognitive performance, most researchers apply the mini-mental state examination (Folstein, Folstein et al. 1975) or Wechsler Adult Intelligence Scale (Dorken and Greenbloom 1953) for evaluation. Chang *et al*. (Chang, Lee et al. 2009) used the clinical dementia rating scale (Morris 1997) to evaluate the functional capability of these patients since they may have physical disabilities. Tasks that have been used for evaluation are as follows: Alzheimer's Disease Assessment Scale-Cognitive wordrecognition test (Rosen, Mohs et al. 1984) for verbal episodic memory; recollection of Rey-Osterrieth complex figures for visuospatial ability (Boone 2000); Boston naming test for language ability (Boone 2000); digit span, digit-symbol, digit backward (Cronholm and Viding 1956; Sherman and Blatt 1968; Rudel and Denckla 1974); Trail Making Part A and Part B, block design, and design fluency (Gieseking, Lubin et al. 1956; Arbuthnott and Frank 2000) for executive function; and neuropsychiatric inventory for behavioral changes (Cummings, Mega et al. 1994).
