Background
Clozapine (CLZ) is considered to be the gold standard in the treatment of refractory schizophrenia. There is a known high risk of mental state destabilisation in patients who have had a 48 h treatment break. The literature on clozapine titrations, including various local guidelines, focuses on initial titration and retitration over a non-standardised duration (SrT), and there is a paucity of evidence examining rapid retitration (RrT) (Nielsen Reference Nielsen, Damkier and Lublin2011; Ronaldson Reference Ronaldson, Fitzgerald and Taylor2012; Ifteni Reference Ifteni, Nielsen and Burtea2014a,Reference Ifteni, Correll and Nielsenb; Medicines.ie 2016; Poyraz Reference Poyraz, Özdemir and Sağlam2016; Taylor Reference Taylor, Barnes and Young2016) (in this article we use RrT to denote a retitration period shorter than 7 days). The available evidence on RrT is from in-patient rather than community settings (Ronaldson Reference Ronaldson, Fitzgerald and Taylor2012; Ifteni Reference Ifteni, Correll and Nielsen2014b; Poyraz Reference Poyraz, Özdemir and Sağlam2016).
Could this lack of evidence be due to the estimated 3% incidence of myocarditis in the Australian population (Ronaldson Reference Ronaldson, Fitzgerald and McNeil2015), which dissuades clinicians worldwide from conducting trials on CLZ RrT because of the perceived high risk?. It has been argued that the risk in the non-Australian population may in fact be as low as 0.07–0.6 per 1000 (Cohen Reference Cohen, Bogers and van Dijk2012). In a Finnish 5-year follow-up of patients with first-onset schizophrenia, the adjusted odds ratio (OR) for cardiovascular death associated with CLZ was found to be 0.23 (95% CI 0.05–1.02) (Kiviniemi Reference Kiviniemi, Suvisaari and Koivumaa-Honkanen2013); and in a Swedish schizophrenia cohort study, CLZ showed an adjusted OR for all deaths of 0.92 (95% CI 0.70–1.22) (Leon Reference Leon, Tang and Baptista2015). In the USA between 1989 and 1999, there were 17 confirmed cases of myocarditis among a total of 189 405 persons treated with CLZ (Grenade Reference Grenade, Graham and Trontell2001). Another retrospective cohort study observed a risk of cardiac mortality, mainly due to myocarditis and cardiomyopathy, in schizophrenia of 1.1% and 2.7% at 5- and 10-year follow-up in people aged under 55, with a further significantly higher risk in the those aged 55 and over (Kelly Reference Kelly, McMahon and Liu2010).
Ronaldson et al (Reference Ronaldson, Fitzgerald and Taylor2012) demonstrated that initial titrations in which cumulative doses above 920 mg were given in the first 9 days of treatment were associated with more than twice the risk of myocarditis in comparison with cumulative doses of less than 500 mg. Each additional 250 mg resulted in an added risk of 26% (OR 1.26; 95% CI 1.02–1.55; P = 0.03). Concomitant use of valproate, older age and change in the patient's smoking status were significantly correlated with risk. Ronaldson et al (Reference Ronaldson, Fitzgerald and Taylor2011) quantified the threshold limits of troponin T (TropT) (at least twice the upper limit of normal) or C-reactive protein (CRP) (>100 mg/L) beyond which CLZ discontinuation should be considered in order to prevent cardiac damage. Compliance with their suggested monitoring protocol may be formidably challenging in community mental health settings.
Ifteni et al (Reference Ifteni, Nielsen and Burtea2014a) evaluated an RrT method (n = 111, average CLZ dose during first 24 h: 25–400 mg) for hospital in-patients, in which no patients experienced seizures, syncope due to hypotension, agranulocytosis or any other major complication. A retrospective study (Poyraz Reference Poyraz, Özdemir and Sağlam2016), also conducted in an in-patient setting, comparing RrT with SrT regimes, found an association of the former with shorter hospital stays, albeit with higher rates of hypotension.
Here we reflect (with the patient's consent) on our clinical experience of RrT and its proactive delivery in a community outreach service for the homeless, where patients are seen according to severity and endurance of illness, up to several times a day. Our service places emphasis on patients with low levels of engagement and a high risk of causing harm to self and others owing to mental destabilisation, through planned community visits.
Case presentation
This 54-year-old homeless man has a 30-year history of paranoid schizophrenia complicated by polysubstance misuse. He has an unremarkable medical history despite his ‘dual diagnoses’. He has a history of self-harm, drug overdoses and impulsive dangerous behaviour. Over the course of his illness he has incurred over 30 mental health admissions. Since commencing CLZ in 2013, a significant clinical improvement, enhanced well-being and risk management have been noted.
He is currently prescribed 450 mg of CLZ daily. Owing to a chaotic lifestyle, often too disorganised to allow him to adhere to treatment, 48 h treatment breaks with clinical deterioration have been frequently noted. To manage these, a 4-day RrT protocol was designed based on the history of his clinical presentation. For treatment breaks longer than 48 h but less than 96 h, CLZ was to be retitrated in single daily doses as follows: day 1, 100 mg; day 2, 200 mg; day 3, 300 mg; day 4, 450 mg. Minimum monitoring requirements for this protocol were blood pressure (BP), heart rate (HR) and temperature, measured before each dose of CLZ. In addition, the risk of myocarditis would be assessed using TropT and CRP tests (as available), with results accessible within hours of requesting, thus allowing prompt changes to the treatment.
CLZ was retitrated on 24 June (RrT1), 31 July (RrT2), 26 August (RrT3) and 1 November 2017 (RrT4) after 48 h treatment breaks, each time following the above protocol. The patient's temperature never exceeded 37.1°C. The highest recorded BP was 140/77 mmHg on day 3 of RrT1 and the lowest was 100/65 mmHg on day 3 of RrT3 (mean BP = 120/72 mmHg, n = 12). The highest recorded HR was 99 bpm on day 3 of RrT3 (mean HR = 90, n = 12). Postural hypotension was suspected in RrT1 and RrT2, and the patient experienced transient syncope on day 1 of RrT1 only. Seven days after initiation of RrT3 and RrT4, TropT and CRP levels were obtained. Although the TropT result was negative (<14 ng/L), CRP levels were elevated (90.72 mg/L after RrT3 and 19.69 mg/L after RrT4 where a normal range would be 0–10mg/L). In RrT3 only, the CRP test was repeated 7 days later (14 days after initiation) and then again 19 days after; these two tests showed a marked reduction (63.34 mg/L and 9.62 mg/L respectively). No changes were made to the patient's treatment, as he was otherwise asymptomatic. Clinical observation indicated that he was under the influence of multiple illicit substances on all four occasions. Overall, there was no destabilisation in his mental state. He did not have any hospital admissions in this period.
Conclusions
We safely and rapidly retitrated CLZ for a high-risk and ‘chaotic’ patient in a community setting four times in 6 months, each time over 4 days. To ensure safety and prevent major complications, we put in place basic monitoring. This case shows that RrT can be delivered even to a patient with severe and complex difficulties. We are now more confident in offering rapid CLZ retitration to other patients. RrT of CLZ in the community can prevent admissions to hospital.
eLetters
No eLetters have been published for this article.