Key messages

  1. 1.

    PKN is the most common described MD secondary to CNZ/FNZ treatment.

  2. 2.

    CNZ/FNZ mechanism of action is the blockage of dopamine, histamine, serotonin, and intracellular calcium-calmodulin complex.

  3. 3.

    All MD can be explained by the dopaminergic hypothesis, except MCL that is probably associated with serotonin.

Introduction

Cinnarizine (CNZ) and flunarizine (FNZ) belong to the calcium channel blockers family (Fig. 1). In this context, CNZ was first synthesized in 1955 by chemistries of the Janssen Pharmaceutica in attempts to develop a new antihistamine drug [1]; animal studies showed that CNZ inhibits vascular smooth muscle contraction, and this action occurs especially in the intracranial vessels [2]. So the idea of a drug with “cerebral vasodilation” properties was evolving [3]; this was probably even more encouraged due to the increasing knowledge about the cerebrovascular diseases [4]. Apparently, a safety profile was observed, and CNZ was marketed in the middle 1960s [1].

Fig. 1
figure 1

Skeletal formula of cinnarizine and flunarizine

Full size image

FNZ was discovered in the same pharmaceutical company as CNZ in 1961; it was designed to have a better dosage form and increase the effectiveness of its dubious predecessor [5]. In animal studies, FNZ showed an efficacy of 2.5 to 15 times stronger than CNZ [6]. It was first marketed in Europe around the 1980s about 10 years after the release of CNZ [3]. Spain among the European countries was one of the first in which FNZ was available [6] and probably the country with the greatest number of CNZ/FNZ prescriptions; in 1985, approximately 5% of the Spanish population over sixty were on CNZ treatment [3].

CNZ/FNZ has been used for several conditions including central and peripheral vascular disorders and balance disorders. Migraine, Raynaud syndrome, Ménière’s disease, vertigo, and tinnitus are examples of common indications of these drugs. The mechanism of action of CNZ/FNZ is poorly understood (Fig. 2) [2, 6,7,8,9,10,11]. These medications were first described as calcium channel blockers with activity in the L/T-type channels [6], but more recent studies showed that the main action of CNZ/FNZ may not be inhibiting calcium entry into cells, but rather by an intracellular mechanism such as antagonism of calmodulin [2, 11]. It is worthy of mentioning that this mechanism is postulated to be effective for the treatment of vertigo [2]. Other theorized interactions of this drug include the blockage of H1 histamine, 5-HT2c serotonin, and D2 dopamine receptors [6, 10]. The serotoninergic mechanism is still poorly understood, some studies showed an increase [7] and in others a decrease [10] of the serotonin concentration in the synaptic cleft [6]. Thereby, we believe that these contradictory results probably occurred because different parts of the brain were studied.

Fig. 2
figure 2

Schematic diagram of the mechanism of action of cinnarizine and flunarizine (CFNZ). CFNZ can act in the vascular smooth muscle blocking the complex calcium-calmodulin (CaM), which will inhibit muscle contraction. In the central nervous system, CFNZ has six main actions; blockage of dopamine D2, histamine H1, and serotonin receptors. Also, it can inhibit the tyrosine hydroxylase and CaM decreasing the synthesis and release of dopamine. Moreover, depending on the site of action, CFNZ can increase serotonin concentration inhibiting serotonin reuptake (SERT) and facilitating serotonin release

Full size image

The side effects observed in the first clinical studies of this drug in more than 1% of the population were drowsiness, nausea, indigestion, weight gain, feeling tired, stomachache, vomiting, sweating, and skin rashes [6]. Other adverse events were only reported in the postmarketing experience [12]. Some patients showed symptoms similar to Parkinson’s disease such as bradykinesia and resting tremors [13]. Later, many reports of drug-induced parkinsonism and other abnormal movements were reported with CNZ/FNZ [13,14,15], which sometimes are difficult to diagnosis in the clinical practice due to preexisting neurological and psychiatric comorbidities. In this way, the aim of this literature review is to evaluate the clinical epidemiological profile, pathological mechanisms, and management of CNZ- and FNZ-associated movement disorders.

Methods

Search strategy

We searched six databases in an attempt to locate any and all existing reports on movement disorders (MD) secondary to CNZ and FNZ published between 1980 and 2019 in electronic form. Excerpta Medica (Embase), Google Scholar, Latin American and Caribbean Health Sciences Literature (Lilacs), MEDLINE, Scientific Electronic Library Online (Scielo), and ScienceDirect were searched. Search terms were “parkinsonism, dyskinesia, dystonia, stuttering, myoclonus, restless legs syndrome, akathisia, tremor, chorea, tics, restlessness, ataxia, ballism, hyperkinetic, hypokinetic, bradykinesia, movement disorder.” These terms were combined with “cinnarizine, flunarizine” (Other 1 - Supplementary material).

Inclusion and exclusion criteria

Case reports, case series, original articles, letters to the editor, bulletins, and poster presentations published from 1980 to 2019 were included in this review with no language restriction. The authors independently screened the titles and abstracts of all papers found from the initial search. Disagreements between the authors were resolved through discussion.

Cases where the cause of MD was already known and either motor symptoms did not worsen or were not related to CNZ/FNZ were excluded. Also, cases that were not accessible by electronic methods including after a formal request to the authors of the study by email were excluded. Cases that had more than one factor contributing to the MD were evaluated based on the probability of the event occurrence based on the Naranjo algorithm.

Data extraction

For CNZ/FNZ, a total of 4032 papers were found; 3431 were irrelevant and 484 were unrelated to the complication, duplicate, inaccessible electronically, or provided insufficient data (Fig. 3). Data abstraction was done. When provided, we extracted author, department, year of publication, country of origin, number of patients affected, CNZ/FNZ indication including off-label uses, time from first CNZ/FNZ dose till MD onset, time from CNZ/FNZ withdrawal to symptoms improvement, patient’s status at follow-up, and important findings of clinical history and management. The majority of the reports did not provide sufficient information about the times of MD onset and recovery. The data were extracted by two independent authors, double-checked to ensure matching, and organized by whether the MD was a side effect of the CNZ/FNZ use.

Fig. 3
figure 3

Flow chart of the screening process for cinnarizine and flunarizine

Full size image

Statistical analysis

Categorical variables were represented as proportions; continuous variables were represented as mean, standard deviation (SD), median, and range.

Definitions

The clinical characteristics and definitions of the MDs such as parkinsonism, tics, dyskinesia, dystonia, myoclonus, restless legs syndrome, akathisia, tremor, chorea, ataxia, and ballism were obtained from the reference Jankovic and Tolosa [16]. The clinical diagnosis for the psychiatric conditions was obtained from the diagnostic and statistical manual of mental disorders (DSM-5®) [17]. The Naranjo algorithm was used for determining the likelihood of whether an adverse drug reaction was actually due to the drug rather than the result of other factors [18]. In the cases where the non-English literature was beyond the authors’ proficiency (English, Portuguese, Spanish, Italian, French, and German) and the English abstract did not provide enough data such as Japanese, Korean, Chinese, Russian, and Dutch, Google Translate service was used [19].

Results

For the years 1980 and 2019, a total of 117 reports containing 1920 individuals who developed a movement disorder associated with CNZ/FNZ were identified from 27 countries (Table 1) [3,4,5, 12,13,14,15, 20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129]. The origin was Asian in 834, European 663, South America 415, and North America 8. Figure 4 shows the number of reports associated with movement disorders and CNZ/FNZ over time with important markers of the history of the CNZ/FNZ-induced parkinsonism [13, 15, 53, 73, 79, 87, 115, 130]. The movement disorders encountered were 1251 parkinsonism, 23 dyskinesias, 11 akathisia, 16 dystonia, and 5 myoclonus. In the group not clearly defined, 592 were extrapyramidal symptoms, 19 tremors, 2 bradykinesia, and 1 myokymia.

Table 1 Clinical reports of CNZ/FNZ-associated MD
Full size table
Fig. 4
figure 4

Graphic showing the number of clinical reports of cinnarizine and flunarizine from 1980 to 2019. The numbers represent important markers of the history of the CNZ/FNZ-induced parkinsonism. (1) De Melo-Souza, a Brazilian neurologist, described the first to report FNZ-induced PKN. (2) Chouza and colleagues published the first report of FNZ-induced PKN. (3) Gimenez-Roldan and colleagues reported that essential tremor and older age were possible risk factors for CNZ-induced PKN. (4) Negrotti and Calzetti studied individuals with FNZ-induced PKN who had been followed for 7 years showing a bad prognosis. (5) Marti-Masso and Poza studied individuals with FNZ-induced PKN who had been followed for 10 years showing a relatively good prognosis. (6) Vazquez-Alen and colleagues showed a large decrease in drug-induced PKN; they hypothesized it was because of the reduction of CNZ/FNZ prescription. (7) Teive and colleagues called the FNZ and CNZ-induced PKN as De Melo e Souza’s syndrome. (8) Population-based studies using Taiwan and Korean databases

Full size image

The resume data about CNZ- and FNZ-associated movement disorders is provided in Table 2. Herein, we will describe the general data of all clearly defined cases.

Table 2 Resume of CNZ/FNZ-associated movement disorders
Full size table

The predominant sex was female with a percentage of 72.69% (466/641). The mean and median age was 74.49 (SD, 7.88) and 71.1 years (age range, 2.5–93 years). The mean and median CNZ dose was 148.19 (SD, 42.51) and 154.4 mg (CNZ dose range, 25–225 mg) and for the FNZ dose, 11.22 (5.39) and 10 mg (FNZ dose range, 5–60).

The mean time from the CNZ/FNZ start to the MD onset was 1.83 years (SD, 1.35). About 75% of the individual had abnormal movement within 3 years of the CNZ/FNZ treatment. The mean time from the CNZ/FNZ withdrawal until the MD recovery was 3.71 months (SD, 1.26). In the subgroup of subjects that had improvement of the symptoms, the complete recovery was achieved within 6 months of the drug withdrawal in almost all subjects (99%). Figure 5 shows a comparison between the percentage of patients developing a MD since the beginning of the drug and the percentage of patients recovering after drug withdrawal when outliers were removed.

Fig. 5
figure 5

Comparison between the percentage of patients developing movement disorders (MD) since the beginning of the drug and the percentage of patients recovering after drug withdrawal. The data is from 663 (MD onset) and 473 (MD recovery) individuals reported in the literature

Full size image

The most common management was drug withdrawal. Other drugs prescribed after the CNZ/FNZ withdrawal included levodopa, anticholinergics (biperiden, trihexyphenidyl, methixene, orphenadrine), benzodiazepines, propranolol, diphenhydramine, and bromocriptine. In individuals that depression was observed, amitriptyline and imipramine were one of the medications started. A complete recovery was observed in 93.77% of the patients (437/466).

Discussion

General

Movement disorders (MD) associated with CNZ/FNZ were commonly reported in the literature. Historical facts probably have contributed to these findings such as the common sense about CNZ/FNZ be always in the list of the drugs that induced parkinsonism [131]; it was the second most common, only after antipsychotics, between the end of the 1980s and early 2000s [87]. Second, the wide number of CNZ/FNZ prescriptions all over the world because of the placebo drugs and the effect of “cerebral vasodilators” [3]. In this way, the well-known side effect and a large number of users’ mixture may explain some of the reports.

Based on the data available in Table 1, we can hypothetically illustrate a case. An elderly Asian female presented with symptoms of vestibular vertigo to her general practitioner. The physician started flunarizine 10 mg. In the long-term follow-up, within about 3 years of the beginning of FNZ, she complained of slow movements, stiffness, and resting tremor. She was diagnosed with drug-induced parkinsonism, and FNZ was withdrawn. Within less than 6 months, the patient had a full recovery.

The number of reports with FNZ was more than 60% of the overall data. Two characteristics of the metabolism of FNZ, when compared to CNZ, that can explain this are the long half-life, which is more than ten times the CNZ, and the accumulation in the central nervous system that is due to the fluorination; FNZ is much more lipophilic than CNZ [3, 6]. Moreover, we believe that another important aspect was the marketing issues with the general major availability of flunarizine all over the world [3].

The majority of the incidences of abnormal movements associated with CNZ/FNZ were not well described in the literature. Table 3 is a resume of the percentages of MD secondary to CNZ/FNZ; the data was extracted from clinical trials and other population-based studies [26, 34, 48, 49, 55, 70, 76, 77, 97, 115, 119, 123, 127, 129]. The incidences of CNZ/FNZ-induced abnormal movements, in general, vary throughout the literature, but the range is much smaller than other drugs with postmarketing evaluation such as valproic acid [132]; for example, extrapyramidal symptoms were found with the use of CNZ/FNZ in Verspeelt and colleagues in 4.30% of individuals with migraine individuals, but in the vestibular vertigo subgroup it was 0.91% [77].

Table 3 Incidence of some abnormal movements in the literature
Full size table

Herein, we would like to discuss some of the MD in subtopics to give a better comprehension of the data.

Parkinsonism (PKN)

History (Fig. 4)

The first report of FNZ-induced PKN was by the Brazilian neurologist De Melo-Souza during the IX Brazilian Congress of Neurology in 1984 [13]. His description of five elderly females who presented PKN and depression was a crucial observation for the knowledge of drug-induced MD [131]. Even though today, the number of cases has decreased and, in many countries, it has restricted prescriptions [2]; FNZ is a widely recognized drug as a cause of secondary abnormal movements and can be an example, as well as antipsychotics, for the MD. Nowadays, this association has been called De Melo e Souza’s syndrome by some authors [130]. It is worthy of mentioning that Martí-Massó and colleagues in 1985 published the first report of CNZ-induced PKN [14].

Incidence

The incidence of PKN with CNZ/FNZ is scarce in the CNZ/FNZ label is that the occurrence is in 1 every 1000 users [6]; in the literature from data extraction of clinical trials, it is 0.07–6.52% (Table 3). It is noteworthy that the prescription of CNZ/FNZ should be avoided in PD. Martí-Massó and colleagues reported that 40% of PD patients with short-term CNZ/FNZ use showed a severe worsening of the bradykinesia and gait [26].

Recent Asian population-based studies showed important features of the long-term CNZ/FNZ use and its complications [115, 119, 122, 124, 127,128,129]. Lin et al. [115] revealed that age, stroke, and diabetes mellitus are risk factors for the development of CNZ/FNZ-induced PKN. Kim et al. [128] showed that the odds ratio of the risk of developing PKN in FNZ users when compared to non-FNZ users is 4.95 (2.71–9.03). In another study, Lin et al. [129] revealed that a longer duration of exposure to FNZ and high FNZ doses are significantly associated with the occurrence of PKN.

Epidemiology and diagnosis

Among the CNZ/FNZ-induced MD, PKN was the most frequently described corresponding for more than half of the cases. The majority of the individuals affected were European, and mainly from Spain probably due to a large number of prescriptions and older European populations [3]. Three epidemiological findings of this subgroup are comparable to the drug-induced PKN of the literature. First, the high incidence in females that is believed to be more susceptible [21], or we can presuppose that this was an occasional finding because females are related to a higher number of prescriptions. The results of Lin et al. that the female sex was not significantly associated with CNZ/FNZ treatment can support this hypothesis [115]. Second, the CNZ/FNZ-induced PKN occurred with higher CNZ/FNZ doses, which may be explained by higher doses leading to higher concentration in the central nervous system and a possible predisposition for the development of this MD [3, 6]. Third, an elderly population was more involved in this subgroup than in other abnormal movements that could be related to aging causing striatum abnormalities [9]; also, the possibility of FNZ/CNZ provoking PD cannot be ruled out.

The presentation in the majority of the cases was a symmetric akinetic-rigid syndrome, with resting and/or postural tremor; in almost half of the individuals, depression (mild, moderate, and severe) was described. Other commonly associated MD in descending order of frequency were akathisia, dyskinesia (orofacial, rabbit syndrome, choreoathetotic), and dystonia (oromandibular). Moreover, Mattos and colleagues reported a patient presenting with progressive supranuclear palsy like syndrome [102].

Sometimes, it can be hard to clearly distinguish between the CNZ/FNZ-induced MD and the idiopathic PD based only on clinical criteria. Thus, Teive et al. [131] selected some clinical tools from the studies of Negrotti and Calzetti [79] and Martí-Massó and Poza [73] to help on the diagnosis of this syndrome. Table 4 has the features by Teive et al. [131], and we propose a supporting feature that is the presence of another MD at the presentation; since an important percentage of the individuals is commonly affected by another disorder.

Table 4 Clinical tools for the diagnosis of CNZ/FNZ-induced PKN by Teive et al. [131] modified by Rissardo and Caprara
Full size table

Pathophysiological mechanism

It is still not completely understood, but some authors believe that it is due to the decrease of dopaminergic neurotransmission [131]. In animal models, CNZ/FNZ decreased the concentration levels of dopamine probably due to tyrosine hydroxylase inhibition and dopaminergic neuron loss [8, 10]; also, the blockage of striatal dopaminergic receptors was observed [9]. We hypothesized that the calcium-calmodulin complex inhibition by CNZ/FNZ may be involved with the decrease of dopamine [11]; the involvement of only the release of dopamine without affecting its concentration or the noradrenaline/adrenaline concentration can support this assumption [10]. Furthermore, in the literature, studies have shown a decrease [10] and/or increase [7] in serotonin concentration with CNZ/FNZ; so, the serotoninergic hypothesis of PKN caused by a mechanism similar to that observed with serotonin reuptake inhibitors could have occurred in susceptible individuals [133].

Management

The most frequent management was the drug withdrawal and the rechallenge was not attempted in any of the cases. Some authors recommended the use of anticholinergics to accelerate the recovery and decrease the number of complications [42, 47]. A full recovery was observed in more than 90% of the cases.

Dyskinesia (DKN)

DKN was the second most commonly encountered MD secondary to CNZ/FNZ. More than 80% of the cases occurred in countries of South America. This can be explained by the great knowledge of this association leading to a possible more minute observation of the clinical findings [131]. The more frequent affected individuals were females 8 years younger than those developing PKN. Also, it was three times more common with FNZ than CNZ.

The presentation more frequently was orofacial. Rabbit syndrome was observed [15], which is an extrapyramidal adverse effect of antipsychotic medicines that perioral tremors occur at a rate of 4–5 Hz; we included in DKN, but there is controversy in the literature and some authors believed that is a separated disorder, which goes beyond the aim of this review. Gabellini and colleagues reported isolated tongue tremors, which was rarely observed in the cases reported in the literature [39].

The most common management was drug withdrawal. However, among the CNZ/FNZ-induced MD, DKN associated with CNZ/FNZ had the worst prognosis with a full recovery obtained in only 66% of the individuals.

One of the possible explanations for the occurrence of this MD is the dopaminergic hypothesis, which happens due to an abnormal adaptation of the striatal organization leading to overactivation of the direct pathway [134]. This hypothesis is plausible in the cases reported and can be supported by the long onset time longer than the average. Moreover, the interaction of CNZ/FNZ with the histaminergic neurotransmission could have contributed to the development of this abnormal movement [6]. The H1 histamine receptors are commonly found in the tuberomammillary nucleus that has many connections with the cerebral cortex, neostriatum, hypothalamus, hippocampus, and nucleus accumbens [135]. Therefore, we believe that the long-term use of the medication can explain some of the cases, in a similar way to other antihistaminic drugs.

Dystonia (DTN)

DTN was observed in 16 individuals, and more than half was reported by South American authors. Some features of this subgroup that are commonly found in the drug-induced DTN literature include the prevalence of female sex predominance, affected younger population (compared to general data), low CNZ/FNZ doses associated with DTN, and short MD onset. The presentation in descending order of frequency was blepharospasm, oromandibular, and cervical; even worsening of a previous DTN was observed in a child with Sturge-Weber syndrome, when FNZ 5 mg was prescribed [126].

The dopaminergic hypothesis can explain the occurrence of CNZ/FNZ-induced DTN. The finding that antipsychotics that also interact with D2 dopamine receptors and are associated with DTN can support this assumption [134]. It has been suggested that the blockage of these receptors in the caudate, putamen, and globus pallidus is partly responsible for causing this abnormal movement [136]. Therefore, it is probably the disbalance of ratio dopamine-acetylcholine especially in the striatum that can produce these symptoms [9].

The most common management was drug withdrawal; in one case, diphenhydramine was started [121]. The complete recovery was noted in 91.66% of the patients.

Akathisia (AKT)

The majority of the reports associated with AKT occurred with higher FNZ doses that were almost twice the mean data. This MD most commonly occurred in the female sex from South America origin. Interesting, AKT was the only MD associated with CNZ/FNZ that after the management 100% of the users had recovered, which we believed that probably occurred because the population affected was 10 years younger than the other MD. The most frequent management was drug withdrawal. Micheli and colleagues reported the first case of CNZ/FNZ-induced AKT; a middle-age male showed AKT symptoms after the first dose of CNZ 75 mg; the drug was withdrawn and the individual recovered in one day [28].

Since the CNZ/FNZ is involved with the dopaminergic system; this hypothesis can feasibly explain all extrapyramidal symptoms. Therefore, as well as PKN, DKN, and DTN, the D2 dopamine block is probably related to the occurrence of AKT [134].

Myoclonus (MCL)

This MD was rarely reported in the literature in association with CNZ/FNZ. Turner and colleagues reported a case of CNZ overdose in a child, she developed twitching in both hands [95]. However, they did not provide a clear description of the neurological examination neither of the electrodiagnostic studies. In another case series, Lopez-Castellanos and Lopez-Contreras reported four subjects in 2017 at the 1st Pan American Parkinson’s Disease and Movement Disorders Congress [120]. The individuals were analyzed by a movement disorder specialist but no description of electromyography or electroencephalogram was done. The presentation was multifocal with or without a tremor. The management was drug withdrawal. Only one individual did not recover after 5 years of follow-up. Based on these two reports, we cannot conclude the source of MCL.

In the literature about drug-induced MCL, the most common hypothesis is associated with the serotoninergic neurotransmission. This abnormal movement was already reported with the increase and the decrease of serotonin concentration [137]. In this context, CNZ/FNZ was first believed to decrease the serotonin concentration, in the synaptic, by the induction of monoaminergic neuron damage probably because the higher number of reports with depressive symptoms in the clinical practice [10], but some studies have shown a contradictory increase of the serotonin that happened due to serotonin reuptake blockage and facilitation of its release [7]. We believed that these different results could have occurred due to different brain sites being studied. Therefore, the increase of this neurotransmitter may be related to MCL in susceptible individuals.

Conclusion

In sum, CNZ/FNZ-associated movement disorders were extensively reported in the literature probably due to important historical features. The most frequent and well-described MD was PKN. MCL was the poorest described MD with missing data about the neurological examination and electrodiagnostic studies. In descending order of frequency, the following MD related to CNZ/FNZ were encountered: PKN > DKN > DTN > AKT > MCL. Most of the CNZ/FNZ-induced MD can be explained by the dopaminergic hypothesis, except MCL that is probably associated with serotonin. We believe that the knowledge of the abnormal movements associated with CNZ/FNZ could significantly raise the awareness of the potential motor side effects secondary to other commonly prescribed drugs. In this way, the continuum development of new drugs with fewer or less severe side effects is essential for the improvement of the quality of life, reduction of negative outcomes, and increase of patients’ adherence.