Early Weight Gain on Psychotropics a Useful Predictor?

Daniel M. Keller, PhD

March 05, 2014

MUNICH — Early weight gain following initiation of atypical antipsychotics or other psychotropic medications may provide a simple, clinically relevant marker of risk for long-term gain and underscores the importance of early weight monitoring, new research suggests.

A 1-year prospective study that included several psychotropic drugs showed that weight gain of 5% or more at 1 month predicted further weight gain at 3 and 12 months.

"It is really important to define an early weight gain cutoff," first author Frederik Vandenberghe, a pharmacist finishing his PhD in the Center for Psychiatric Neurosciences at Lausanne University Hospital in Switzerland, told delegates attending the 22nd European Congress of Psychiatry (EPA).

To that end, the investigators sought to calculate the best predictive power of weight gain at 1 month on long-term weight gain. Patients were monitored at baseline and at 1, 2, 3, 6, and 12 months. A switch in treatment or drug discontinuation was considered a dropout. Medical staff assessed compliance, with therapeutic drug monitoring at 1, 3, and 12 months.

Of the 315 patients, 47% were male. The average age of the patients was 46 years, with ages ranging widely from 15 to 96 years, but the durations of the patients' illnesses were fairly close at 8 ± 0.6 years. Baseline body mass index was 24.4 ± 0.3 kg/m2. Mean follow-up duration was 236 ± 8.2 days.

Researchers calculated sensitivity and specificity for weight gains of 1% to 8% at 1 month as predictors of amounts of gain ranging from 5% to 20% at subsequent intervals.

On the basis of area under the curve reflecting the best sensitivity and specificity, they selected a cutoff of 5% at 1 month as the best predictor of weight gain at 3 months.

Vandenberghe noted that his hospital has had a metabolic monitoring program that has been looking at second-generation antipsychotic drugs as well as lithium, valproate, and mirtazapine since 2007.


Table. Predictive Power of 5% Weight Gain at 1 Month

Area under the curve: 79
>5% gain at 1 month Prediction of >15% gain at 3 months
Sensitivity: 67% Specificity: 88%
Positive predictive value: 29% Negative predictive value: 97%
Area under the curve: 68
>5% gain at 1 month Prediction of >20% gain at 12 months
Sensitivity: 47% Specificity: 89%
Positive predictive value: 35% Negative predictive value: 93%


Men and women who gained >5% at 1 month experienced a weight gain of about 12% at 3 months and a weight gain of about 14% at 1 year compared with baseline. Those patients who gained ≤5% at 1 month had a 3-month gain of only about 2% and a 1-year gain of around 5%.

The amount of weight gain at 1 month predicted the degree of worsening of metabolic parameters at 12 months. Looking at 12-month values vs baseline values, glucose levels (P = .02) and triglyceride levels (P = .004) rose more, and high-density cholesterol levels (P < .001) decreased more in the higher weight gain group when compared with those who had gained <5% at 1 month.

Vandenberghe concluded that >5% weight gain during the first month of treatment is a good predictor for further weight gain at 3 and 12 months. He suggested that this threshold should be a warning sign to reconsider treatment options and to advise patients to adopt weight loss strategies.

One audience member asked whether higher weight gain at 1 month was predictive at all age groups, especially in the elderly. Vandenberghe replied that when his group stratified patients by age, the early excessive weight gain still predicted more weight gain later. Even for patients older than 65 years, "we saw the difference remained significant," he said.

Another audience member asked whether it was ethical in the study not to implement measures to limit weight gain in those patients who were over the 5% threshold at 1 month. He replied that the study was observational, and furthermore, patients whose medications were switched were dropped from consideration in the study. So patients' weight gains were not ignored, he added.

A Good Marker?

Session chair Aye Mu Myint, MD, PhD, of the Department of Psychiatry at Ludwig Maximilian University in Munich, Germany, who was not involved with the study, agreed that weight gain is not safe in young or old patients. But the question is whether interventions will work.

"Systematic studies should be done [on] what type of interventions, and different patients might have different factors that induce them to get this weight gain [and] metabolic syndrome," she told Medscape Medical News.

"Possible individual factors may include inflammation, glucose tolerance, and insulin receptor sensitivities. I think we should do interventions, but in a personalized medicine approach," said Dr. Myint.

Weight gain of <5% had a high negative predictive value (ie, specificity) but a low positive predictive value. If people do not gain more than 5% at 1 month, then they will probably not gain excessive weight later and get metabolic syndrome.

"But if they gain weight [of more than] 5%, it doesn't mean that then they will then get metabolic syndrome," Dr. Myint explained. "So just using that, you can't do intervention because it is not really sensitive.... I would not suggest this marker is very good for the patients because sensitivity is important, too."

In essence, the weight gain at 1 month can tell who does not need intervention but is not particularly good at saying who is very likely to gain further weight.

Frederik Vandenberghe and Dr. Myint report no relevant financial relationships. There was no commercial funding for the study.

22nd European Congress of Psychiatry (EPA). Session FC08. Presented March 4, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.