Obesity, characterized by the excess accumulation of fat in body, is a chronic disease which is induced by many causes and needs long-term management. Obesity prevails all over the world. Moreover, obesity leads to the increase of several co-morbidities and all-cause mortality, and produces extreme economic losses for management of the obesity-related diseases.1-4 The situation is same in Korea. The obese population [body mass index (BMI) ≥ 25 kg/m²] rate of Korea has soared by 12.6%, from 23.3% in 1992 to 35.9% in 2000.5

Several methods have been tried in order to prevent and treat the increasing rate of obesity and its related disorders. One of such attempt is the application of appetite-suppressing drugs, which is used in many countries for reducing the energy intake. Among these drugs, phentermine (Ionamin®) is the drug which had been approved for short-term use as treatment of obesity by US FDA in 1959, and has been used steadily until present. Phentermine, classified as a β-phenethylamine drug, stimulates the secretion of noradrenalin in central nervous system, and noradrenalin, in turn, suppresses appetite by stimulating the β-adrenergic receptors.6,7

One of the reasons why phentermine is still prescribed in the era of appearance of several new anti-obesity drugs is that it has been administered safely without serious side effects for the past 40 years in US. However, because of the fact that phentermine is a sympathomimetic amine, we should keep in mind the possible side effects of the drug of this class in using phentermine. Although the combination of fenfluramin and phentermine have been demonstrating a remarkable weight reduction effect, US FDA has removed fenfluramin from the market in September 1997, due to the successive reports about the serious side effects of valvular heart disease and primary pulmonary hypertension. There have been scarce reports about serious side effects for the single use of phentermine; however, there is a researcher who argued that 15 of 1000 phentermine administered patients might experience serious side effects.4

Another reason for wide distribution of phentermine prescription is its low economic burden for use in contrast to the new drugs. Comparing the cost-effectiveness of sibutramine and phentermine, in US, 103.8$ per month for sibutramine 10 mg daily induces 4.45 kg weight reduction in 12 months. On the other hand, 39.59$ per month for phentermine resin 30 mg daily induces 3.6 kg weight reduction in 6 months.8 Such data regarding this cost-effectiveness might be the important criteria for the selection of which drug to use.

Phentermine has been introduced in Korea since 2004, and have been prescribed widely. The reports of the randomized controlled studies of phentermine had been published from 1975 to 1999, and there have been no further clinical trials on the phentermine usage since 1999.9 Moreover, there are few reports about the effect and safety profile of phentermine in Korean. Therefore, we performed a double-blind placebo-controlled trial of phentermine in order to reaffirm the effect of phentermine on weight reduction and to verify the safety in Korean obese patients.

In total, 94 men and women have displayed intention to participate in the study. 15 of them were eliminated as they violated the initial instruction during screening period, and 11 were eliminated as they fell in with the exclusion criteria. Actual number of participation subjects was 68, and 13 men and 55 women were randomly assigned to double-blind treatment. 35 were randomized to phentermine group, and 33 to placebo group. 36 of 68 (52.9%) had completed the 14-week treatment course. 24 of 35 phentermine group (68.6%), and 12 of 33 (36.4%) had completed the study.

At baseline, there was no significant difference between the groups in terms of age, sex, weight, waist, blood pressure, and smoking rate (Table 1).

During the two-week run-in period, the mean decrease in weight was 0.58 (SD 1.18) kg, and that of BMI was 0.22 (SD 0.46) kg/m², with the associated reduction of 1.20 (SD 3.19) cm in waist circumferences across both groups.

In the analysis of the ITT population, there was a statistically significantly greater mean reduction in both body weight and waist circumference in phentermine-treated subjects to subjects in placebo group (p < 0.001) (Fig. 1, Table 2). There was no significant difference in blood pressure change between the two groups after the treatment.

Table 3 shows the result in completers. In the analysis of completers, there was also a statistically significantly greater mean reduction in both body weight and waist circumference in phentermine-treated subjects to subjects in placebo group (p < 0.001) (Table 3). Including the weight reduction of the run-in period, placebo group shows about 2 - 3 kg of weight reduction and phentermine group shows about 7 kg of weight reduction (Table 2, 3). Run-in period subtracted analysis showed that phentermine HCl 37.5 mg was associated with significant weight reduction (-7.1 [SD 2.4] kg for completers and -6.7 [SD 2.5] kg for ITT) and shortening of waist circumference (-6.9 [SD 3.3] cm for completers and -6.2 [SD 3.5] cm for ITT; data not shown in table; all p < 0.001).

In the analysis of both ITT and completers, a significantly greater proportion of patients in the phentermine group achieved weight reduction of 5% or greater from the baseline compared with the placebo group, and it was same for the subjects who lost 10% or more weight from baseline (Fig. 2).

In completers, the changes of blood pressure, lipid levels, fasting plasma glucose, high-sensitive CRP and uric acid from baseline to endpoints are shown in Table 3. There was no statistically significant difference between the groups in these variables except total cholesterol. After the treatment, the mean change of total cholesterol in phentermine group was -7.8 (SD 28.5) mg/dL, compared to 10.7 (SD 20.2) mg/dL in placebo group. The changes from baseline to endpoints in each group were not significantly different from zero. However, the difference between groups was significant (p = 0.048, Table 3).

In addition, the change of non-HDL-cholesterol between groups was also significant (data not shown in table, p = 0.023 by Wilcoxon rank sum test). The mean change of non-HDL-cholesterol in phentermine group was -9.7 (SD 24.8) mg/dL, while the mean change in placebo group was 7.8 (SD 24.8) mg/dL. However, the changes in each group were not significant, same as total cholesterol.

The proportion of patients who had reported any side effect was higher than that of patients who had not reported. Table 4 provides an analysis of all the adverse events occurred in at least 5% of patients in any group. The adverse events in Table 4 are listed in the frequency order of phentermine group. The number of subjects who had reported any adverse events in phentermine group was significantly larger than that in placebo group. However, the reported rate of adverse events in placebo group was even with 75% (18/24). Among all reported adverse events, dry mouth and insomnia are the events which occurred significantly more frequently in phentermine group. The occurrence rates of the other adverse events were not significantly different between the groups. For the most part, these events were mild to moderate in intensity, and only a few cases needed to reduce the dosage.

There were two subjects who withdrew due to adverse events. One of them withdrew because of dry mouth and foreign body sensation in throat. The other withdrew due to nausea and headache. Another two subjects complained severe adverse symptoms. One complained of severe headache, and the other, of severe headache and nausea. However, after reducing dosage by half, their symptoms had improved much or totally removed, and both of them had succeeded in completing the study. All four were included in phentermine group.

The discontinuation rate was significantly higher in placebo group (p = 0.008), and there was significant difference in the mean duration of study participation between two groups, 11.0 (SD 4.9) weeks for phentermine group vs. 8.2 (SD 5.3) weeks for placebo group (p = 0.030). In phentermine group, 11 of 35 (31.4%) withdrew after run-in period. Seven of them discontinued study participation without any clear reason, and two of them discontinued because of adverse events. One subject withdrew due to no weight losing effect of the drug. There was one subject who discontinued due to the difficulty in following the behavioral guideline. In placebo group, 21 of 33 (63.6%) withdrew after run-in period. 15 of them discontinued study participation without any clear reason. There was nobody who withdrew because of adverse events. Five was due to personal schedule, and one due to no weight losing effect of the drug. Among 114 reports of total adverse events in subjects who were treated for more than 6 weeks, only 31 (27.2%) had newly occurred after the 6th week of treatment, and 24 (21.1%) symptoms had prolonged for more than 4 weeks. In phentermine group, 13 of 29 had reported the occurrence of adverse events after the 6th week of treatment, and in placebo group, 4 of 24 had reported after the 6th week.

In completers, there was no significant difference of change of level from baseline to endpoints in RBC, WBC, and platelet counts, hemoglobin, hematocrit, BUN, creatinine, total bilirubin, AST and ALT levels.

Generally, currently recommended obesity treatment method is the improvement of life style, the essentially featuring reduction of calorie intake and increasing physical activity. However, the outcome of such method have shown to be disappointing. Therefore, several drugs have been developed for weight reduction, and a few of them had been approved for obesity treatment. Phentermine is one of these drugs. It is a drug which had been approved by US FDA in 1959 with a brand name of Ionamin®, and more than 50 million prescriptions of which have been written out since 1960. Despite its wide usage, few clinical trials of phentermine for the weight reducing effect and safety have been performed in Korea. Phentermine is produced in two major forms. One is phentermine resin (eg. Ionamin®) and the other is phentermine-HCl (eg. Adipex®), where the latter is released more rapidly.

Cost is an important factor in the pharmacologic therapy of obesity. The difference of cost-effectiveness between sibutramine and phentermine in US has already been described in introduction, which is guite the same in Korea. Moreover, obesity treatment is not covered by medical insurance in Korea until present in 2006. Therefore, total drug purchasing expense for obesity treatment is burdened to patients. Until 2006, Reductil® and Xenical® are the only drugs approved for long-term treatment of obesity in Korea. Both of these drugs are expensive. Purchasing cost for 1 month as initial recommended dosage is over 100,000 won in 2006. However, the price for the use of phentermines is around one fourth of the price for using Reductil® or Xenical®. Relatively lower cost of phentermine maybe of assistance for those patients who have difficulty in purchasing highly priced obesity drugs.

In this study, 2 weeks run-in period and 12 weeks administration of phentermine-HCl 37.5 mg had induced clinically significant weight reduction, shortening of waist circumference, and reduction of total cholesterol and non-HDL-cholesterol level. This result would reduce the risk of cardiovascular disease in the ultimate. Moreover, over 80% of subjects of phentermine group lost 5% or more of initial weight and more than half subjects lost 10% or more. This shows us that most of obese patients can obtain the conventional goal of obesity treatment by short-term use of phentermine.

There have been many reports about the weight reduction effect of phentermine until today. In 1968, Munro et al.14 performed long-term double blind placebo-controlled study on 108 obese women with phentermine resin. They compared three groups of placebo, continuous phentermine therapy and intermittent phentermine therapy (administration of phentermine and placebo every 4 weeks, alternatively) for 36 weeks, and reported that both continuous therapy (mean weight loss 12.2 kg) and intermittent therapy (13.0 kg) resulted in significant weight reduction than placebo (4.8 kg, p < 0.001). In this study, they reported that phentermine had held the effect of gradual loss in weight even after the 6 months of drug administration, which is different to other weight-losing drugs Langlois et al.15 performed placebo-controlled study on 59 patients with phentermine hydrochloride for 22 weeks (the duration of drug administration was 14 weeks) in 1968. In this study, phentermine group (mean weight loss 16.1 kg) showed significantly more weight loss than placebo (3.9 kg, p < 0.001). In both studies, experiments reported that side effects were not serious. Weintraub et al.7 reported the result of placebo controlled study in which they compared the effect of phentermine resin 30 mg, fenfluramine hydrochloride 60 mg and phen-fen (phentermine resin 15 mg + fenfluramine hydrochloride 30 mg) on 81 subjects for 24 weeks in 1984. In this result, phentermine group (mean weight loss 10.0 kg), fenfluramine group (7.5 kg), and phen-fen group (8.4 kg) showed significantly superior weight reduction effect than placebo. However, phentermine only group induced more side effects than phen-fen combination therapy or placebo. In addition to these studies, there had been several other small sized clinical trials using phentermine, however, all of them showed similar weight reduction effect, and reported similar side effect profiles which could be mimicked with by subjects in general.

Desirably, the process of weight reduction should be taken gradually and continue for a long time. However, this is difficult. Therefore, the intentional weight losing process should be divided into two periods, initial weight reduction period and sustaining reduced weight period. The guideline of US National Institutes of Health16 recommends that the target weight which is 10% or more less than initial weight should be gained over 6 months, and then the patient should attempt to maintain it. It is because, after certain amount of weight reduction, it becomes very difficult to lose more weight due to several adaptation processes to reduced weight such as decreased basal metabolic rate. In our study, phentermine group gained 7.2 kg of weight loss on average for 14 weeks, and this amount of reduced weight is 9.3% of initial weight. The result of our study shows that phentermine can induce initial target weight, even though the clinical trial of long-term usage of phentermine is needed.

Phentermine is classified as sympathetic amine, which shows its effect by the secretion of norepinephrine at nerve terminal. However, the exact appetite suppressing mechanism of phentermine is not fully understood. Several studies on this topic suggest that phentermine may suppress appetite on hypothalamus by the increase of the concentration of norepinephrine,17 dopamine18,19 and serotonin20 in CNS. Phentermine usually induce tachycardia and increase blood pressure because of its sympathomimetic effect. In this study, pulse rate had not regularly checked on every visit. However, in adverse events analysis, only 6 of 29 (20.7%) phentermine group subjects complaint on their palpitation, and this number is not significantly more than placebo group. Similarly, blood pressure had not increased after treatment period in phentermine group, and there had been no significant difference between both groups. Not only the blood pressure of endpoints but also the blood pressure of 4 weeks administration of phentermine is not significantly different between two groups. Not shown in the results, the analysis of blood pressure at 6th week, i.e. after 4 weeks administration of phentermine, the mean systolic pressure was 127.4 (SD 10.2) mmHg and the mean diastolic pressure was 85.1 (SD 8.4) mmHg in phentermine group (n = 28), compared to systolic 127.3 (SD 12.7) mmHg and diastolic 83.5 (SD 10.4) mmHg in placebo group (n = 24). Therefore, we can suggest that the administration of phentermine would not induce any serious cardiovascular effect at least for the patients whose risk of cardiovascular disease is not high.

In our study, the administration of phentermine caused just a few serious adverse events. In completers, no subjects showed abnormal result on standard laboratory test. Although the number of subjects who had reported adverse events was much larger than that of subjects without adverse events, and more subjects in phentermine group complained on side effects than subjects in placebo group, dry mouth and insomnia were the only adverse events that occurred significantly more frequently in phentermine group. Moreover, these events were not so serious as to prevent the subjects from participating in the study. Drinking plenty of water could reduce dry mouth symptom and in most cases, it resolved of itself. 5 of 28 phentermine completers reported that they had experience dry mouth during the whole period of phentermine administration. There was nobody who reported that their insomnia prolonged over 4 weeks. In many cases of insomnia, by switching the administration time to the morning the symptom could be improved.

According to the direction of Gate pharmaceuticals which is one of drug companies producing phentermine drugs, known adverse effects of phentermine are primary pulmonary hypertension, valvular heart disease, palpitation, tachycardia, elevated blood pressure, hypersensitivity, dizziness, insomnia, mood change, tremor, headache, driness of the mouth, unpleasant taste, diarrhea, urticaria, impotence, and changes in libido, etc.21 At the initial briefing on the research procedure to the subjects, the possibility and symptoms of primary pulmonary hypertension and valvular heart disease were informed and the subjects had been instructed to notify to the researchers immediately if they had felt related symptoms. However, there had been no report of symptoms related to these two serious adverse events in the study period. The majority of other known adverse effects had been reported in the study. For taste change, there was one report in placebo group and the symptom was minimal and self-limited. There was no report of taste change in phentermine group. There was no report of tremor and diarrhea in both groups. There had been no definite report of urticaria. However, one subject reported minimal temporary itching sensation at the 10th week in phentermine group, and another reported temporary skin lesion on face at the 6th week which could not been identified by researchers because of complete recovery on the subsequent clinical visit. There had been no report on impotence and change of sexual desire, and this might be due to the self report method of side effects whichever they felt during study participation, not the selection method on side effects list.

There had been seven reports about epigastric pain and chest discomfort. After detailed history taking and physical examination, all of these reports had been considered to be not related to cardiac problems, and therefore no further diagnostic study had been performed. In the study, the subjects were relatively healthy obese people; the age of the subjects was not high and there had been no subjects with hypertension and diabetes mellitus, both of which are strong risk factors of cardiovascular disease. Only 23.5% of subjects were smokers, and the rate was relatively lower than known average smoking rate amongst Korean. It might be due to the female abundance of the subjects. Therefore, the possibility of cardiovascular disease in this study group might be much lower than that of average Korean obese population. In real practice of obese patient with administration of phentermine, the health provider should pay attention to the symptoms of epigastric pain or chest discomfort especially with the high risk patients with cardiovascular disease. Then, again, further evaluation on the cardiac related symptoms could be performed even though for the case of non cardiac disease. Such aspect of phentermine administration might be the limitation of usage on the obese patients with high age, diabetes mellitus, or past medical history of cardiovascular disease.

Phentermine is classified by US Drug Enforcement Agency (DEA) as schedule IV drug.22 The completers had been requested to visit the researchers at the 18th week for the check up of withdrawal symptoms or other unexpected adverse events, at where they were provided with laboratory results. 12 of 24 (50%) phentermine completers and 6 of 12 (50%) placebo completers visited researchers. There had been nobody who complained withdrawal symptoms or other adverse events.

Phentermine hydrochloride is indicated by US FDA as a short-term (a few weeks) adjusted in a regimen of weight reduction based on exercise, behavioral modification and caloric restriction in the management of exogenous obesity for patients with an initial BMI ≥ 30 kg/m², or ≥ 27 kg/m² in the presence of other risk factors (e.g., hypertension, diabetes, and hyperlipidemia). A short-term is generally accepted as around three months. Some experts insist that if the initial three month effect on weight reduction is acceptable good, the prolonged administration of phentermine will be appropriate.23 This study is for the evaluation of the effect and the safety of short-term usage of phentermine. The clinical trials for long-term effect and safety are also necessary.

Conclusively, we can suggest that short-term phentermine administration can induce significant weight reduction and the shortening of waist circumference without clinically problematic adverse events on relatively healthy obese people, by this study.