We regularly see comments that Sumatriptan changes the experience of CH--increasing frequency and/or intensity. What is usually missing are any data on frequency of this experience, duration of changes, source of the claim, and so on. Several years ago I searched medical literature for some specifics on this experience and could only find the two abstracts (below). In January, 2011 I searched for more current reports and could not find anything in the previous 10-years.

My conclusion is: the absence of later data suggests that there is little experience stimulating reports/study and that this is not an important problem. The last point is reinforced by the observation that in the few reported events, that the changes in headache reversed when Sumatriptan was dropped.

As with other medical topics, it's important that we qualify our "truth" claims with parameters/limits which don't distort real life experience.

It's not much help to people to warn them off using a good treatment with a silent implication that some side effect is widespread, enduring, even dangerous. We are always, with every medical treatment, struggling to balance benefit vs. risk. As we expect our physicians to fairly present the pros & cons of a treatment/procedure to us, we should, given our limited knowledge & skills, try to do the same.
==================================

Headache. 2000 Jan;40(1):41-4.

Alteration in nature of cluster headache during subcutaneous administration of sumatriptan.
Hering-Hanit R.

Headache Unit, Department of Neurology, Meir General Hospital, Kfar Sava, and the Sackler Faculty of Medicine, Tel Aviv University, Israel.

Abstract
OBJECTIVES: To document the relationship between the 5-HT receptor agonist sumatriptan and a change in the nature of cluster headache in four cases. To relate the findings to the literature on the use of sumatriptan in both cluster headache and migraine.

BACKGROUND: Studies of the efficacy and adverse effects of long-term treatment with sumatriptan in cluster headache are limited and report conflicting findings.

METHODS: FOUR CASES ARE DESCRIBED.

RESULTS: All four patients developed a marked increase in the frequency of attacks 3 to 4 weeks after initiating treatment with the drug for the first time. Three patients also developed a change in headache character, and 2 experienced prolongation of the cluster headache period. WITHDRAWAL OF THE DRUG REDUCED THE FREQUENCY OF HEADACHES AND ELIMINATED THE NEWLY DEVELOPED TYPE OF HEADACHE.

CONCLUSIONS: Determination of the effects of long-term use of sumatriptan will result in more precise guidelines for the frequency and duration of treatment with this otherwise extremely beneficial drug.

PMID: 10759902 [PubMed -
=======================================

Headache. 2004 Jul-Aug;44(7):713-8.

Subcutaneous sumatriptan induces changes in frequency pattern in cluster headache patients.
Rossi P, Lorenzo GD, Formisano R, Buzzi MG.

Headache Centre, INI Grottaferrata, Rome, Italy.

Comment in:

Headache. 2005 Sep;45(8):1089-90.

Abstract
OBJECTIVES: To document the relationship between the use of subcutaneous (SQ) sumatriptan (sum) and a change in frequency pattern of cluster headache (CH) in six patients. To discuss the clinical and pathophysiological implications of this observation in the context of available literature.

BACKGROUND: Treatment with SQ sum may cause an increase in attack frequency of CH but data from literature are scant and controversial.

METHODS: Six CH sum-naïve patients (three episodic and three chronic according to the International Headache Society (IHS) criteria) are described.

RESULTS: ALL SIX PATIENTS had very fast relief from pain and accompanying symptoms from the drug but they developed an increase in attack frequency soon after using SQ sum. IN ALL PATIENTS, THE CH RETURNED TO ITS USUAL FREQUENCY WITHIN A FEW DAYS AFTER SQ SUM WAS WITHDRAWN OR REPLACED WITH OTHER DRUGS. Five patients were not taking any prophylactic treatment and SQ sum was the only drug prescribed to treat their headache.

CONCLUSIONS: Physicians should recognize the possibility that treatment of CH with SQ sum may be associated with an increased frequency of headache attacks.

PMID: 15209695 [PubMed

What makes research in this area is that it is hard to isolate cause and effect with the very small sample sizes available. For the two studies quoted by Bob, one had 4 subjects and the other 6. Most of us have variations in quantity, severity and duration of our CHs, with potentially more variation for those who are episodic as it's possible that given the experiment duration it's likely that some of those who were episodic were either near the start or the end of a cycle.

Another factor that could strongly influence things is the methodology used where the test was done on people who were using sumatriptan for the first time. It's possible that this does not correlate with the experience of those who have been using sumatriptan for a long duration. A simple example is when you see the behaviour of people starting smoking for the first time compared to those who have been using it for a long time. Or compare people first eating very spicy food to those who eat it regularly.

Neither experiment reported the use of controls, where some subjects used sumatriptan and some didn't. Ideally experiments should be double blind, with neither the subject or the person interacting with the subject knowing if the subject was being given sumatriptan or a control, e.g. water. Although since there is a very obvious difference in effects, using controls is probably not possible with sumatriptan given the very obvious effects of sumatriptan on a CH.

Doing experiments where there is a natural random variation with a very small sample will always be difficult with limited data sets resulting in relatively poor statistical data sets to analyse.

Ricardo: I think we agree that there is scant evidence of a problem but that's not the issue I'm trying to surface.

Problem, as I see it, are folks who make absolute claims of a problem when, there is scant to no evidence to support their claim.

It's an ethical issue that's the heart of my message.