Imipramine versus placebo for multiple functional somatic syndromes

[Murph]

Imipramine versus placebo for multiple functional somatic syndromes (STreSS-3): a double-blind, randomised study.

https://www.ncbi.nlm.nih.gov/pubmed/28408193

Abstract
BACKGROUND:
Functional somatic syndromes, including chronic fatigue syndrome or irritable bowel syndrome, often co-exist. Treatment guidelines supported by high quality evidence exist for most functional somatic syndromes, but are lacking for multiple comorbid functional somatic syndromes. We aimed to assess the effect of the tricyclic antidepressant, imipramine, in patients with multiple functional somatic syndromes defined by the criteria for multiorgan bodily distress syndrome, a unifying diagnosis that encompasses most functional somatic syndromes and somatoform disorders.

METHODS:
In this single-centre, double-blind, randomised trial done in a Danish university hospital setting, participants were patients consecutively referred (age 20-50 years) fulfilling criteria for multiorgan bodily distress syndrome with no concurrent comorbid depression or anxiety disorder. Participants were randomly assigned (1:1) to receive either 10 weeks of low-dose imipramine or placebo (oral daily doses of 25-75 mg). The hospital pharmacy handled randomisation (computer-generated) and masking, providing sequentially numbered packs of study drug that were given serially to the participants.

All others involved were masked to allocation. Primary outcome was patient-rated overall health improvement on a 5-point clinical global improvement scale. Improvement was defined as patients responding "better" or "much better" as opposed to "unchanged" and "worse" or "much worse" when rating their overall health status after 10 weeks of minimum 25 mg study drug. Analyses included patients who received at least one dose of study drug. This study is registered with ClinicalTrials.gov, number NCT01518634.

FINDINGS:
Between Jan 30, 2012, and Nov 24, 2014, 138 patients were randomly assigned; 70 to receive imipramine and 68 to receive placebo. The study was completed on May 1, 2015. 125 patients received at least one dose of study drug: 65 received imipramine and 60 received placebo. Treatment was terminated prematurely for eight (12%) patients receiving imipramine and seven (12%) patients receiving placebo. Data were missing for two (3%) patients receiving imipramine and three (5%) patients receiving placebo. Of the 120 patients (96%) who provided primary outcome data, 33 (53%) receiving imipramine reported their overall health status as "better" or "much better" compared with 14 patients (25%) receiving placebo.

The improvement after imipramine was significantly greater than after placebo (odds ratio 3·3 [95% CI 1·6-6·8]; p=0·001). Number needed to treat was 3·6 (95% CI 2·3-8·9). Analysis of the worst-case scenario for patients with missing outcome did not change the interpretation of the results. 32 patients (49%) receiving imipramine and 10 patients (17%) receiving placebo had at least one adverse event of moderate intensity (p=0·0001); eight patients (12%) receiving imipramine and three patients (5%) receiving placebo had at least one adverse event of severe intensity (p=0·1496). One patient (1%) receiving placebo experienced a serious adverse event (a subdural haematoma sustained after an accident). Adverse events caused dropout in four patients (6%) receiving imipramine and three patients (5%) receiving placebo.

INTERPRETATION:
Imipramine treatment compared with placebo significantly improved overall health in patients with multiple functional somatic syndromes when both treatments were supported by regular contacts with clinicians. Adverse events were more common in the imipramine group, but only rarely led to discontinuation of treatment.

nb. According to this Fink, CFS is included in "bodily distress syndrome."

 

[halcyon]

So..doesn't that basically prove that people meeting the bodily distress syndrome criteria have an organic illness? If they were truly functional somatic syndromes, shouldn't the placebo group have improved as well?

 

[trishrhymes]

Since a low dose of tricyclics (eg amitriptiline) is a common, well established symptomatic treatment for pain and sleep problems in all sorts of conditions, surely all they were showing was that it helps symptomatic relief in some ME and IBS patients too.

Given the fluctuating nature of ME and IBS, it's not surprising that 25% of the placebo group were 'improved' at the end of trial.

Despite being properly double blinded etc, this trial seems pretty poor to me - small numbers of patients, including patients in the results who had taken 'at least one dose' of their allocated medication, and quite a high rate of adverse events, and no mention of numbers who got worse or used other treatments at the same time.

I have only read the abstract, so don't have full information on how they measured improvement. Just using a global 'are you feeling better' kind of measure seems very crude and limited. Why not detailed assessment of actual IBS symptoms, and actometer etc. for ME symptoms? Ah, but I was forgetting - these are people classing us as having functional somatic sydromes, not real illnesses.

 

[TiredSam]

Imipramine versus placebo for multiple functional somatic syndromes (STreSS-3): a double-blind, randomised study.

This is beyond satire. A double-blind, randomised study is for studying things that actually exist. Trumpeting the method to lend credibility to a delusional belief system (multiple functional somatic syndromes) is laughable. How about "Garlic versus wooden stakes for monstrous blood-sucking syndrome: a double-blind, randomised vampirical study."

 

[Effi]

Agger JL1, Schröder A2, Gormsen LK3, Jensen JS2, Jensen TS4, Fink PK2.

 

[Dolphin]

They mention the lack of objective measures in the limitations section:

Our trial has several limitations, one being the choice of outcomes. Both the primary and all secondary outcomes relied on patients’ self-report with no additional clinician-rated or objective measurements of illness severity or functioning (eg, walking distance, work ability). The effect of imipramine in these domains remains unclear. However, objective measurements are generally lacking in functional somatic syndromes research, which makes our study comparable to other similar trials.29 Symptom scores are often proposed as the most reliable outcome measures in functional somatic syndromes. These measures have the inherent disadvantages of any symptom score system—ie, they measure specific symptoms that might have no effect on the patients’ global wellbeing. In studies of fibromyalgia, improvement in symptom score corresponds well with improvement measured by means of global scores.20 Furthermore, global scores have the advantage of not being specific to certain functional somatic syndromes. We argue that patient rated overall health improvement, our choice of primary outcome, reflect the most important endpoint in multiorgan bodily distress syndrome, especially considering the fluctuation in symptom location and severity and the wide and varied symptomatology of multiorgan bodily distress syndrome.

I think it is less of an issue in a double-blind trial like this.

 

[Dolphin]

Despite being properly double blinded etc, this trial seems pretty poor to me - small numbers of patients, including patients in the results who had taken 'at least one dose' of their allocated medication, and quite a high rate of adverse events, and no mention of numbers who got worse or used other treatments at the same time.

They do give the percentage of who reported themselves worse in the full text:

Imipramine (n=65) Placebo (n=60)

Much worse 1 (2%) 1 (2%)
Worse 4 (6%) 11 (18%)
Unchanged 25 (39%) 31 (52%)
Better 22 (34%) 14 (23%)
Much better 11 (17%) 0 (0%)
Missing 2 (3%) 3 (5%)

Data are number (%) in response to the question: “How do you consider your health status now compared with when you first came to the clinic?”.

Table 2: Primary outcome—Clincal Global Improvement raw score

 

[Dolphin]

I thought it was interesting that 0% in the placebo group reported being "much better".

23% in the placebo group did report being better, some of which could be due to biased reporting due to placebo and some might be due to random fluctuations of their illness.

 

[Dolphin]

It would have been interesting to have the magnitude of the improvements for the secondary outcomes. We are just given the odds ratios compared to the placebo group.

 

[Dolphin]

Primary reasons for exclusion hereafter were psychiatric disorders demanding treatment or ongoing pain medication.

Up to eight tablets of 500 mg paracetamol were available daily as escape analgesia.

Patients undergoing concomitant treatment with antidepressants, anticonvulsants, analgesics, or other medication with pain relieving properties were excluded, unless this medication could be discontinued.

 

[Dolphin]

The health improvement after imipramine was not exclusively attributable to pain relief, because those patients who did not report pain as their predominant symptom also had an overall health improvement after imipramine. The health improvement seems to be not attributable to change in mental health either, because neither the mental health scales of the SF-36 nor symptoms of anxiety or depression improved after imipramine compared with placebo. This lack of differences in mental health between patients receiving imipramine and placebo allows us to conclude that the health improvement after imipramine in this study is not caused by an indirect effect of change in mental health— eg, a relief in an underlying, subclinical depression.

 

[Dolphin]

Imipramine (n=65) Placebo (n=60)

Overall adverse events*

No adverse events 7 (11%) 17 (28%)
At least one adverse event of at least moderate intensity† 36 (55%) 12 (20%)
Patient drop-out due to adverse event 4 (6%) 3 (5%)

On the other hand, along with a possibly increased susceptibility, the patients seem to possess both robustness and stamina when experiencing symptoms. Despite the adverse events, treatment was continued by most patients.

I thought the table with the specific adverse events was interesting as it showed that the drug did tend to cause more adverse events in a number of different domains/symptom areas.

 

[Dolphin]

Apart from the comments above, I don't have major criticisms of the trial.

A related tricyclic drug, trimipramine (Surmontil, Rhotrimine, Stangyl) has helped me with sleep and pain though my overall functioning remains low.

 

[Dolphin]

Imipramine (n=65) Placebo (n=60)

Functional somatic syndromes†

Chronic fatigue 52 (80%) 48 (81%)
Fibromyalgia 43 (66%) 44 (75%)
Irritable bowel syndrome 20 (31%) 23 (39%)
Non-cardiac chest pain 25 (39%) 35 (59%)
Tension headache 48 (74%) 40 (67·8%)

Number of functional somatic syndromes 3·8 (2·5) 3·8 (2·5)

†Based on functional somatic symptoms in the past 2 years, according to diagnostic interview and review of medical records; data are missing for one (2%) of 60 patients in the placebo group.