Thursday, August 9, 2007

Study criticism: "A cure for dry eye" (and a little editorializing about letters to the editor)

Few things in life give me so much pleasure as truly masterful use of language.

I admire it most when used for a constructive purpose. After all, it's easy and almost commonplace to be amusingly eloquent as a critic. Furthermore, caustic wit is usually directed at people, which I do not like witnessing.

Once in awhile, though, a critique of an issue or performance is so badly needed that when someone steps up to the plate and swings, there is great pleasure to be had in seeing the ball not merely struck square on and flying out of the park but making heavenly music as it soars. (Particularly pleasureable when contrasted to the worn-out tired old sports metaphors I slip into when I'm on a no-coffee kick.)

Anyway, this morning I experienced a mouth-wateringly beautiful specimen of this type of writing. It comes from the unlikely pen of a pediatric ophthalmologist.

Sandra Brown is a master of the genre of the Letter to the Editor of a peer-reviewed medical journal. Now, when I say master, I mean The Real Thing. I am not talking about just effectively pointing out the flaws of a study. I am not talking about unusually skillful manipulation of fact and language to prove a truly important point. I'm not even talking about the exercise of genius in content, style, wit and timing in addition to all the above. I'm talking about a breathtakingly concise musical arrangement, an articulation of exclusively relevant facts wielding linguistic scalpels with such deftness and delicacy that the artistry of it makes you laugh for joy. Dr. Brown has taken this craft to a level of unequalled sophistication.

But the beautiful part is that it's not the people who are targeted. Even those who come in for some collateral damage may remain mercifully untouched by the humiliation since the average reader even of a medical journal probably does not have the insider insights to appreciate each and ever needle - and in any case, the absence of emotion in Dr. Brown's tone seems to take some of the personal sting out of even the deepest of her needle jabs.

No. Dr. Brown's target is, as always, Bad Science. That's Bad Science with a capital B and a capital S.

I'm referring, of course, to "A Cure for Dry Eye" (Ophthalmology, August 2007).

The context: A study published in January's Ophthalmology which claimed that "Topical cyclosporine treatment appears to be associated with a cure of symptoms and signs in a subgroup of chronic dry eye patients".

Dear Editor:

I read with interest the recent article by Wilson and Perry describing 8 patients with dry eye syndrome who were cured "of symptoms and signs" after treatment with topical cyclosporine 0.05% (Restasis, Allergan Inc., Irvine, CA). In their Tables 2 and 3, the authors report objective findings (tear breakup time, Schirmer testing, vital dye staining) before and after treatment; these data support their conclusion that examination findings of dry eye resolved.

Their Table 1 gives patient demographics, including specific pretreatment symptoms. We are not given primary data regarding posttreatment symptoms or lack thereof; the authors declare in narrative that all subjects were symptom-free. However, clinicians know that patients who are miserable will arrive with a symptom list in hand; patients who are much improved and are satisfied with their current treatment may report "all better", which means that they do not feel the need to complain about low-grade and infrequent symptoms.

In general, journal editors and reviewers should require better science in the measurement of dry eye symptoms, now that simple, efficient, validated patient questionnaires are available (e.g. the Ocular Surface Disease Index, developed by Allergan). Despite the authors' national recognition as experts in the field of dry eye, readers are under no obligation to take their word for it that these unusual patients were entirely without symptoms after cyclosporine treatment. Further, the greater the perceived potential for investigator bias, the more carefully authors must defend their neutrality through excellent methodology in data collection and analysis.

My high school English teacher railed against the phrase very unique, as an object or event is either unique or it is not. Similarly, a patient is either improved or cured, and mostly improved is not cured. Cure is a potent word, and I do not believe the authors have proved their statement that topical cyclosporine cures symptoms of dry eye syndrome in a subgroup of patients.

Sandra M. Brown, MD
Concord, North Carolina


Ophthalmology, August 2007, p. 1585-1586

No comments: