Rivastigmine patch for treatment of Alzheimer's disease in clinical practice in Thailand


  • Potential conflict of interest: This study was financially supported by Novartis (Thailand) Ltd. All authors reported no other potential conflicts of interest.

Dr Bandit Thinkhamrop PhD, Department of Biostatistics and Demography, Faculty of Public Health, Khon Kaen University, Muang 40002, Thailand. Email: bandit@kku.ac.th


Background:  Rivastigmine is a cholinesterase inhibitor for treatment of mild to moderate Alzheimer's disease (AD) and dementia associated with Parkinson's disease. The new patch formulation was recently made available. We assessed the safety, tolerability, and cognitive outcome of rivastigmine patch in treatment of mild to moderate AD in clinical practice in Thailand.

Methods:  A multicentre, hospital-based, prospective observational study was conducted in nine hospitals across Thailand. Patients with probable mild to moderate AD who received the rivastigmine patch were enrolled. Data were collected data at baseline, weeks 4–8 and after week16.

Results:  A total of 116 AD patients were screened, and three were excluded. Of 113 patients, 62.8% were women with a mean age of 73.3 ± 9.2 years; 79.7% were newly diagnosed. One-third of all patients had been using antipsychotic or antidepressant medication. Common comorbidities were hypertension and dyslipidemia. The Thai Mental State Examination score significantly increased from 18.6 to 20.3 (weeks 4–8) and 20.4 (week 16+) (P < 0.001). Scores based on physicians’ (Clinical Global Impression) and caregivers’ (Patients’ Caregiver Global Impression of Change) impressions of improvement suggested minimal improvement. Because of adverse events, seven patients's dosages were reduced 10 cm2 to 5 cm2 or from 5 cm2 to nothing. Itching was the most common adverse symptom.

Conclusions:  During the first 16 weeks after initiation of rivastigmine patch therapy, patients with probable mild to moderate AD had statistically significant improvement in cognitive function, but clinically marginal benefit. Rivastigmine was safe and well tolerated.


To date, acetylcholinesterase inhibitor remains a standard treatment for Alzheimer's disease (AD). Rivastigmine is a cholinesterase inhibitor that inhibits both butyrylcholinesterase and acetylcholinesterase, cholinesterase enzymes that would otherwise break down the brain chemical acetylcholine.1 Initially, rivastigmine was developed for oral administration in capsule and solution preparation. Its common side effects are nausea and vomiting because of its short half-life and fluctuation in blood levels. Therefore, a transdermal patch has been developed to improve the pharmacokinetics of this medication. The patch lowers blood maximal concentration with markedly less fluctuation between peak and trough plasma levels. Average exposure with the 10-cm2 patch was comparable to the highest capsule dose (6 mg twice a day, or 12 mg/day).2

Although the efficacy of rivastigmine has been demonstrated in many studies,3–5 few have reported on postmarketing clinical experience with the patch.6,7 Recent postmarketing data from Taiwan and Canada showed improved and/or stabilized cognitive and global functioning.8,9 However, because of lower body weight and other factors, the clinical response may vary in different populations. A pharmacokinetic study in Japanese patients showed slightly higher drug exposure and more pronounced inhibition of butyrylcholinesterase than in Caucasians, which was attributed to the Japanese patients’ lower body weight.10 This may lead to different treatment outcomes. Special attention must be paid to patients with very low body weight during up-titration of the treatment.11 Therefore, we conducted this study to assess the safety, tolerability and clinical outcome of the rivastigmine patch in a real-life clinical practice in Thailand. Secondary objectives included assessing cognitive outcomes of patients using the Thai Mental State Examination (TMSE), Clinical Global Impression (CGI), and Patients’ Caregiver Global Impression of Change (PCGIC).


Patients and study design

This was a multicentre, hospital-based, prospective observational study conducted at nine hospitals across Thailand between July 2009 and June 2011. To be eligible, men and women not of childbearing potential were at least 50 years old, diagnosed with probable AD according to DSM-IV criteria, had a TMSE score between 10 and 26, and received the rivastigmine patch. Patients were excluded if they were involved in other clinical trials or had been treated with an experimental drug within the previous 4 weeks, had an active skin lesion that would prevent accurate assessment of the adhesion and potential skin irritation of the patch, had a severe or unstable physical illness (e.g. acute and severe asthmatic conditions, severe or unstable cardiovascular disorders, active peptic ulcer disease, or clinically significant laboratory abnormalities), had bradycardia (beat per minute less than 50) or a history of arrhythmia, weighed less than 35 kg, and had hypersensitivity to cholinesterase inhibitors.

The medicinal product under investigation was the Exelon patch (Novartis, Basel, Switzerland). There are two doses available in Thailand: 5 cm2 (4.6 mg/24 h) and 10 cm2 (9.5 mg/24 h). Prescriptions of the medication, including dose titrations, are based on real-life clinical practice.

Outcome measures

Patients who met the inclusion/exclusion criteria and their caregivers who consented were enrolled. Patients made three visits to the clinic. At visit 1, baseline characteristics of the patients were obtained, including demographic data, medical history, and past and current medications. We also performed a physical examination (including vital signs and weight), prescribed the study medication, and administered the TMSE. At visits 2 (week 4–8) and 3 (week 16+), patients or caregivers were queried for adverse events and concurrent medications. We also performed a symptom-directed physical examination as required, a general physical examination (including vital signs, weight and questions about concomitant medications), a cognitive assessment using the TMSE, CGI and PCGIC, and a compliance evaluation.

For cognitive assessment, a TMSE score less than 26 was classified as having dementia. For the CGI score, we measured the change from baseline and changes were classified as follows: 1, very much improved; 2, much improved; 3, minimally improved; 4, no change; 5, minimally worse; 6, much worse; and 7, very much worse. For the PCGIC score, we measured the change from baseline, and scores ranged from 0 (much better) to 10 (much worse). For the PCGIC evaluation at visits 2 and 3, caregivers were asked to provide a patient's overall degree of change compared to that person's baseline. The same caregiver was invited to accompany the patient to the clinical visits to ensure there was no change of the caregiver during the study period.

Tolerability was measured as the proportion of patients who dropped reduced their dosage either from 10 cm2 to 5 cm2 or from 5 cm2 to nothing at the follow-up visits. Safety refers to the total number of all episodes of adverse events (including serious adverse events) divided by the total person-weeks under observation. Safety assessments consisted of monitoring and recording all adverse events, including serious adverse events.

Statistical methods

Demographic and baseline variables were described by summary statistics. Percentage distributions were presented for all categorical variables. For continuous variable, mean ± SD and median (minimum: maximum) were used.

The study samples used for analysis of the primary endpoint (i.e. safety and tolerability of rivastigmine) included all patients who received at least one dose of rivastigmine during the study and had at least one safety assessment at a follow-up visit. Patients who had received at least one dose of rivastigmine during the study but who had no safety assessment data of any kind were excluded from analysis.

Proportions of adverse events and tolerability, along with their 95% confidence intervals (95% CI), were estimated based on exact binomial distribution. Mean change from baseline and the TMSE at visit 3 was estimated and tested using the paired t-test. Percentage distributions were compared between the two follow-up visits for CGIC and PCGIC scores.

All analyses were performed using STATA v.10 (StataCorp, College Station, TX, USA). Significance was set as 0.05, and all statistical tests were two-sided.


Patients and participating hospitals

A total of 116 patients diagnosed with probable AD from nine hospitals across Thailand were screened, and three were excluded (Fig. 1). Of the 113 patients enrolled, seven dropped before at the second visit: one because of adverse events and six did not follow up. Ultimately, 106 patients came to visit 2 and 95 patients to visit 3. The rate of loss at follow-up was 6.2% at visit 2 and 15.9% at visit 3.

Figure 1.

Algorithm of the study. wk, week.

Baseline characteristics of the patients

Of 113 patients, 62.8% were women, and 53.3% attained primary school (i.e. 4 years of formal education) (Table 1). Their mean age was 73.3 ± 9.2 years and mean weight was 55.2 ± 9.6 kg. Approximately one-quarter of patients (24.1%) were overweight or obese. Most patients (98.2%) lived with a caregiver, 75.5% were naïve to treatment of AD, and 79.7% were newly diagnosed. Approximately one-third of patients were currently using antipsychotic or antidepressant medications. The most common comorbidities included hypertension (49.6%) and dyslipidaemia (41.6%).

Table 1. Baseline characteristics of the patients
Age at enrolment, mean ± SD (years)73.3 ± 9.2
Sex, women (n (%))64 (62.8)
Duration of formal education, mean ± SD (years)6.0 ± 4.4
Weight, mean ± SD (kg)55.2 ± 9.6
Body mass index (n (%)) 
 Lower than 18.5 (underweight)15 (14.4)
 18.5–24.9 (healthy weight)64 (61.5)
 25.0–29.9 (overweight)21 (20.2)
 30.0 or greater (obese)4 (3.9)
Current living situation (n (%)) 
 Alone2 (1.8)
 With caregiver111 (98.2)
History of Alzheimer's disease treatment (n (%))
 Naïve86 (75.5)
 Previous cholinesterase inhibitor17 (15.3)
 Others8 (7.2)
Reason for switch to rivastigmine patch (n (%))
 Side effect1(6.2)
 Unsatisfied efficacy4 (3.5)
 Poor compliance5 (4.4)
Comorbidities (n (%)) 
 Hypertension56 (49.6)
 Dyslipidaemia47 (41.6)
 Diabetes mellitus24 (21.2)
 Previous stroke8 (7.1)
 Renal insufficiency5 (4.4)
 Ischemic heart disease4 (3.5)
 Parkinson's disease3 (2.7)
 Others (one for each patient)44 (38.9)
Patients who currently used antipsychotic or antidepressant medication (n (%))35 (31.3)
Thai Mental State Examination score, mean ± SD18.6 ± 4.8

Status at follow-up visits

The mean follow-up duration was 15.7 ± 5.7 weeks (Table 2). The living status (i.e. with a caregiver) of most patients (87.6%) remained unchanged at visit 3. Patients’ pulse rates and blood pressures were also unchanged at all visits (data not shown). The proportion of patients who used antipsychotic or antidepressant medications before and after Exelon treatment was similar (31.3% vs 33.6%, respectively).

Table 2. Status of the patients at date of follow-up
Duration of follow-up , mean ± SD (weeks)15.7 ± 5.7
Changes of living status (n (%)) 
 Changes at visit 2 only8 (7.1)
 Changes at visit 3 only5 (4.4)
 Changes at both follow-up visits1 (0.9)
 No change99 (87.6)
Using of antipsychotic or antidepressant medications (n (%))
 Patients who used before Exelon treatment35 (31.3)
 Patients who used after Exelon treatment38 (33.6)

Dose titration patterns and tolerability profile

The most common titration pattern involved patients initiating treatment with rivastigmine patch 5 cm2, then increasing to 10 cm2 during weeks 4–8, and increasing to 10 cm2 again at week 16+. This pattern accounted for 49 patients (46.2%) who made at least one follow-up visit. At the end of the study period, 67 of the remaining patients (70.5%) had initiated the maximal dose of 10 cm2 (Fig. 2). However, seven patients (6.6%) reduced their rivastigmine patch dosages from 10 cm2 to 5 cm2 or from 5 cm2 to nothing because of adverse events.

Figure 2.

Flow of dose titration patterns.

Changes of clinical status

The mean TMSE score at baseline was 18.6 (Fig. 3); it increased significantly to 20.3 at week 16+ (P < 0.001), a mean change of 1.7 (95% CI: 1.1 to 2.7). For physicians’ CGI evaluations, 50% of patients were mainly judged to be minimally improved visit 2 and 39.0% were minimally improved at visit 3 (Fig. 4). More patients were improved at visit 3 (29.2%) than at visit 2 (23.1%) (Fig. 5). However, this change was not statistically significant (P = 0.481). Based on PCGIC evaluations, caregivers perceived not change in 33.3% of patients at visit 2 (Fig. 6). A slightly greater proportion of patients who had a PCGIC score of 4 or less were perceived by caregivers as being better at visit 3 (56.8%) than at visit 2 (47.7%). However, this difference was not statistically significant (P = 0.671).

Figure 3.

Changes of Thai Mental State Examination (TMSE) score from baseline.

Figure 4.

Percentage breakdown of physician evaluations of patients based on the Clinical Global Impression scale at 4–8 and 16+ weeks after treatment.

Figure 5.

Percentage breakdown of physician evaluations of patients based on the Clinical Global Impression that were improved, had no change, or were worse at weeks 4–8 and week 16+.

Figure 6.

Percentage breakdown of caregiver evaluations of patients based on the Patients’ Caregiver Global Impression of Change scale at weeks 4–8 and week16+.

Adverse events

Among 95 patients who remained at the end of the study, 15 (14.2%) had at least one adverse event (95% CI: 08.1–22.3) (Table 3). Itching was the most common symptom, affecting 14.2% and 15.8% of patients at visits 2 and 3, respectively. There was one serious adverse event, diarrhoea, which affected 0.94% of patients (95% CI: 0.02–5.14).

Table 3. Frequency of adverse events and serious adverse events
Adverse events (AE) n (%)
  • 95% CI: 8.1–22.3. 95% CI: 0.02–5.14. SAE, serious adverse event; CI, confidence interval.

Symptoms of AE at the last visit 
 Itching15 (15.8)
 Skin rash4 (4.2)
 Nausea and vomiting3 (3.2)
 Dizziness3 (3.2)
 Contact dermatitis2 (2.1)
 Headache2 (2.1)
 Insomnia2 (2.1)
 Others6 (6.3)
Rate of AE at any visits 
 091 (85.3)
 113 (12.3)
 21 (0.9)
 31 (0.9)
 Total106 (100)
Percentage of all patients (n= 106) who had at least one AE15 (14.2)
Percentage of all patients (n= 106) who had SAE1 (0.9)


This is the first study in Thailand that has assessed the safety and tolerability of rivastigmine patch in AD patients in clinical practice. We measured safety, tolerability, patients’ cognitive outcome, and physicians’ and caregivers’ global impression on patients’ improvement at weeks 4–8 and week 16+. Most patients (79.7%) were newly diagnosed with mild to moderate AD; patients’ mean TMSE score was 18.6 ± 4.8, and they were mainly treated with a titration pattern starting at 5 cm2 (4.6 mg/24 h), increasing to 10 cm2 (9.5 mg/24 h) at weeks 4–8, and then increasing to 10 cm2 again at week 16+. These profiles are consistent with recommendations for rivastigmine patch treatment.12 We stopped follow-up after the third visit, which was approximately 16 weeks after the initial treatment. As such, the duration of the higher dose (10 cm2) was shorter than that of the lower one (5 cm2).

Nonetheless, cognitive outcomes improved from the treatment, which lasted on average 15.8 ± 5.7 weeks, based on changes in TMSE scores. The increase in mean TMSE score from 18.6 at baseline to 20.3 at the last visit of the study, a mean change score of 1.7, is clinically marginal but promising. It should be noted that results from a retrospective analysis of a large randomized controlled trial (IDEAL) of rivastigmine suggested a dose higher than 9.5 mg/24 h to gain substantial clinical benefits.7

CGI and PCGIC scores remained unchanged or minimally improved. However, there were potential confounding factors, specifically the short-term practice effect of TMSE testing and inter-individual variability in patients’ cognitive and functional decline. Given the similar rate of psychotropic drug (antidepressants or antipsychotics) use, the effect of rivastigmine on behavioural or psychiatric outcome was not demonstrated in this study, but this may have been due to short follow-up period and relatively mild disease stage.

Itching was the most common side effect, with approximately 15.8% of patients having this symptom. The discontinuation rate because of these adverse events was 6.6%, or 7 of 106 patients who made at least one follow-up visit. There was one serious adverse event (0.94%), which was diarrhoea. These findings are consistent with previous studies,6,13–16 and these low rates suggest the safety of rivastigmine. Common adverse events such as skin reactions can be avoided by simple daily rotation of the patch location.14 In the IDEAL study, 3.7% of patients discontinued treatment because of application-site skin reactions during 24 weeks.16 In our study, the corresponding proportion was 6.6% during 16 weeks. This reflects real-life practice under uncontrolled settings. Discontinuation could be reduced by following manufacturer's advice about patch application (i.e. daily rotation of the application site) to minimize the risk of skin reactions.16 In addition, providing patients and their caregivers with detailed information about skin reactions could omit unnecessary drop-outs to treatment.6

In our study, 70.5% of patients achieved the maximum dose of 9.5 mg/24 h rivastigmine patch at week 16 compared to 78.2% in the 24-week IDEAL study.4 Among the 63 patients who received the maximum dose at weeks 4–8, 81.0% were able to maintain the dose at week 16, which was slightly higher than what was found in a similar study (74.2%).17

The primary limitation of this study is the short duration of the follow-up period (mean, 15.7 weeks; maximum, 46 weeks). This is because the study focused on tolerability issues during the titration phase, so the initial efficacy and ability to achieve a high-dose patch may suggest a positive long-term adherence and its outcome. However, the findings should not be viewed as a result of long-term treatment.

In conclusion, rivastigmine patch was safe and well tolerated during the first 16 weeks of treatment in a real-life setting in patients with mild to moderate AD who had never been treated with rivastigmine. In addition, the patients had statistically significant improvement in cognitive function but clinically marginal benefit.


The authors would like to thank the Clinical Research Collaboration Network for its coordination and monitoring during the study period. The authors would also like to thank patients and their caregivers for their cooperation.