Discussion

The prescription rate for clozapine is 50% in our FPC. Treatment-resistant schizophrenia (TRS), for which clozapine is the standard recommended treatment, is the most common indication. Clozapine has also been shown to be effective in reducing hostility, aggression and violence in patients with schizophrenia and is of benefit in comorbid substance use disorders.

Psychotic forensic patients often lack insight in their disease and tend not to adhere to antipsychotics. As forensic patients often forcefully resist clozapine treatment and threaten legal or disciplinary action, psychiatrists are more apprehensive about clozapine's potentially lethal side-effects. Development of fever and tachycardia combined with fatigue, muscular pain and weakness in the initial phase of treatment should alert the psychiatrist. Clinical guidelines, such as the Dutch clozapine guideline,¹ provide clinicians with essential information about leukocyte monitoring and the treatment of common (hypersalivation, constipation) and uncommon (myoclonus, hypotension) side-effects of clozapine. When leukopenia has been excluded, myocarditis, NMS and rhabdomyolysis must be considered. Clozapine-associated NMS is a rare condition but must be identified when present.

The enzyme creatine kinase facilitates the transfer of a phosphoryl group from phosphocreatine to ADP, thereby transforming ADP into ATP. This process results in the local production of ATP for the most energy-demanding processes in the human body, such as muscle movement. In healthy adults, plasma creatine kinase consists of three subfractions, the dimers CK-MM, CK-MB, and CK-BB: CK-MM originates in muscle tissue, CK-MB in the myocard and CK-BB in the brain. As CK-MM makes up >97% of total creatine kinase, total creatine kinase in plasma or serum is commonly used for diagnosis of non-cardiac pathology. The activity of creatine kinase is measured by a standard biochemical laboratory test that can be ordered 24/7 in most clinical chemistry laboratories.

In our patient, a sharp increase in creatine kinase levels led to clozapine discontinuation twice. This laboratory finding was wrongly equated with muscle damage, which in turn was interpreted as a foreboding of NMS. The consensus diagnostic criteria for NMS^{Reference Gurrera, Caroff, Cohen, Carroll, de Roos and Francis2} are recent dopamine antagonist exposure or dopamine agonist withdrawal, hyperthermia, rigidity, sympathetic nervous system instability, tachycardia with tachypnoea, and raised creatine kinase levels (four times the upper limit). Historically, raised creatine kinase levels were considered a major manifestation of, if not the essential test for, NMS.^{Reference Levenson3} Creatine kinase levels higher than 1000 IU/L were, in the absence of intramuscularly administered medication, considered suspect for the syndrome or associated organic pathology.^{Reference Manéro Macen, Malo Ocejo, Uriarte Uriarte and Pinilla Tubet4} However, strongly elevated creatine kinase levels (>1000 IU/L in three cases) were measured in 70% of patients (n = 10) who developed fever on psychotropics without NMS.^{Reference O'Dwyer and Shephard5} Clozapine initiation caused creatine kinase elevation without any clinical features of NMS.^{Reference Keshavan, Stecker and Kambhampati6}

In a study comparing patients who did or did not develop NMS, creatine kinase levels were raised in all patients (n = 10) with the syndrome, 60% of whom had levels higher than 1000 IU/L.^{Reference Maniam and Rahman7} However, creatine kinase levels were also raised in 67% of patients (n = 30) without the syndrome (20% >1000 U/L); this was associated with intramuscular injections and/or physical restraint. In a separate study (n = 17), creatine kinase levels that became raised after intramuscular injections returned to normal within 72 h after the last injection.^{Reference Wilhelm, Curtis, Birkett and Kenney Herbert8}

In short, raised creatine kinase levels in the absence of other clinical signs and symptoms tend to be a non-specific finding and do not constitute NMS. Indeed, a patient with fully developed NMS and normal creatine kinase levels has been reported,^{Reference Nisijima and Shioda9} as has a patient with atypical NMS without fever but with raised creatine kinase levels.^{Reference Collins, Davies and Menon10} Not only do the clinical characteristics of NMS with clozapine and other antipsychotics not differ,^{Reference Ananth, Parameswaran, Gunatilake, Burgoyne and Sidhom11} but clozapine has been given successfully to eight out of nine patients who developed NMS on other antipsychotics.^{Reference Weller and Kornhuber12} Atypical presentation with little, no or delayed creatine kinase elevation^{Reference Cherry, Siskind, Spivak, Wysoczanski and Halangoda13} can complicate and delay the diagnosis of NMS.^{Reference Teo, Wong and Tan14} Clozapine and second-generation antipsychotics (SGA) in general are said to induce more varied presentations of the syndrome. The timely and accurate diagnosis of atypical NMS can therefore be a difficult and challenging task for the clinician.

Unfortunately, knowledge about more common and benign creatine kinase elevation has shifted to the background. For instance, primary antipsychotic-induced creatine kinase elevation or massive asymptomatic creatine kinase elevation (MACKE), which is characterised by a transient increase (2–20 times the upper limit) of skeletal muscle creatine kinase without excessive exercise or trauma, was first reported in acutely psychotic patients in 1968.^{Reference Meltzer15} In 10% of patients (n = 121) treated with SGA, massive increases in creatine kinase were observed. They appeared to be more frequent in patients on SGA maintenance, and were higher and persisted longer than in patients treated with first-generation antipsychotics (FGA).^{Reference Meltzer, Cola and Parsa16} Increases above the upper limit were present in 17% of patients on SGA (n = 36) versus none on FGA (n = 13).^{Reference Melkersson17} Selectivity is another point: in one case, creatine kinase increased with olanzapine and quetiapine but not with clozapine.^{Reference Boot and de Haan18} Periodic monitoring of clinical symptoms and creatine kinase levels has been recommended by several studies.^{Reference Boot and de Haan18,Reference Voros, Osvath, Fekete and Tényi19} A systematic review of 42 published cases of MACKE – nine with clozapine treatment – found a prevalence of 2–7%.^{Reference Laoutidis and Kioulos20} MACKE was more frequent in males and in SGA-treated patients. The presence of symptoms was strongly related to myoglobinuria but unrelated to the magnitude of the creatine kinase elevation. The probability of a new creatine kinase elevation was not significantly higher after rechallenge than after switching to another antipsychotic. In the case of clozapine, a rechallenge could be considered, mainly because of the lack of alternatives. Measurement of creatine kinase is advised in patients on antipsychotics who develop muscle symptoms.

Strenuous physical exercise in the previous 24 h can significantly raise creatine kinase levels, which gradually return to normal over the course of a week. The magnitude of the increase depends on the nature and duration of the exercise, and on the training status of the person (with greater increases in the untrained), and there are high inter-individual variations.^{Reference Baumert, Lake, Stewart, Drust and Erskine21} Physical trauma (including epileptic fits or intramuscular injections), hypothyroidism, and rheumatic or neuromuscular diseases can increase creatine kinase levels.^{Reference Duyff, Van de Bosch, Laman, van Loon and Linssen22} Statins are responsible for a 2–10-fold increase in CK levels in about 5% of treated individuals, probably as a result of low-level myopathy or rhabdomyolysis.^{Reference Law and Rudnicka23} Macro creatine kinase and idiopathic elevated serum creatine kinase are less common causes of creatine kinase elevation.^{Reference Moghadam-Kia, Oddis and Aggarwal24} Macro creatine kinase is an abnormal enzyme complex with an atypically high molecular mass (detectable by electrophoresis^{Reference Galarraga, Sinclair, Fahie-Wilson, McCrae, Hull and Ledingham25,Reference Wyness, Hunsaker, La'ulu, Rao and Roberts26} ) and reduced clearance, resulting in abnormally high blood levels of creatine kinase. Macro creatine kinase is present in 4% of patients with asymptomatic or minimally symptomatic elevated creatine kinase. Idiopathic elevated creatine kinase is a persistent creatine kinase elevation with negative findings on examination, testing (electromyography, nerve conduction and muscle biopsy) for a neuromuscular disease and family history for neuromuscular disease.^{Reference Rowland, Wilner, Cerri, Dimauro, Miranda, Angelini, Danielli and Fontanari27}

Although more has been learned about changes in creatine kinase levels in recent years, two findings have not yet percolated down to current clinical psychiatry: the differences between the two sexes and those among races. This observation is explicitly stated in only two of the studies^{Reference Meltzer, Cola and Parsa16,Reference Boot and de Haan18} discussed above. In 1974, creatine kinase activity was shown to be higher in African Americans than in Caucasians.^{Reference Meltzer and Holy28} In 1983, Wong et al^{Reference Wong, Cobb, Umehara, Wolff, Haywood and Greenberg29} found high (52–520 U/L) creatine kinase levels in African American men; intermediate (35–345 U/L) levels in Caucasian men and African American women; and low (25–145 U/L) levels in Caucasian women. The sex and ethnic differences are clear, but these values cannot be used in clinical practice because they date from before assay standardisation and the use of international units. A second study^{Reference Black, Quallich and Gareleck30} confirmed differences in creatine kinase levels between the sexes and between people of Caucasian and African origin, while people of Hispanic origin showed creatine kinase levels in between those of the other two groups. These differences in creatine kinase concentrations were similar in tissue and plasma.^{Reference Brewster, Coronel, Sluiter, Clark and van Montfrans31} The National Health and Nutrition Examination Survey showed significantly larger differences between the 90th, 95th and 97.5th percentiles in Black compared with White males (Table 1).^{Reference George, McGill and Baker32} Evidence suggests a non-Gaussian distribution of creatine kinase values in people of African origin.^{Reference George, McGill and Baker32} Although possible confounders (differences in muscle mass and renal clearance) have been excluded, a satisfactory explanation of this phenomenon is still lacking.

Table 1 Creatine kinase levels in the US population^{Reference George, McGill and Baker32}

To complicate matters further, intra-individual variability in creatine kinase levels in healthy individuals can be as high as 15.4%.^{Reference Aarsand, Fernandez-Calle, Wesbster, Coskun, Gonzales-Lao and Diaz-Garzon33} Correct interpretation of creatine kinase levels requires reference values in male and female persons of different ethnic origins, and repeated measurement of creatine kinase levels over the course of 2–3 weeks.

In Box 1, different causes of creatine kinase elevation are summarised. In our patient, we considered the changes in creatine kinase levels during clozapine treatment to represent normal intra-individual variations for his ethnicity and sex, although benign clozapine-induced primary creatine kinase elevation could not be ruled out.