Editorial: Abdominal binders

Abstract

Sklar and colleagues2 describe their experience managing “over-shunting headaches” with an abdominal binder. Seventy children with over-shunting headaches complied with application of a binder for about 1 month. In 61 patients (87%), the headaches “greatly improved or went away.” This headache relief persisted even after use of the binder was discontinued. Among the 61 patients with relief, 36 (59%) eventually had recurrent headaches, but the recurrence was delayed (mean 1.5 years). Twentynine of these tried the binder again and among the 19 with follow-up, the binder was again effective in 15. These are interesting results. Children with chronic headaches and small ventricles can be very difficult to treat, often undergo repeated surgical interventions, and may have a poor quality of life. Anything that might help them is welcome, especially a simple noninvasive intervention. On the other hand, proving the efficacy of a treatment requires adherence to clinical research methods that are designed to eliminate bias and objectively assess outcome. A list of 6 simple criteria from Sackett et al.1 helps us evaluate clinical papers on therapy. I have applied them below. Was the study design appropriate? No. This was a retrospective chart review, a study design often used as the first step to evaluate a new idea (and sometimes the only step, especially in surgery). There was no control group. The outcome was assessed by the treating (that is, unblinded) surgeon. Despite best intentions, the potential for observer bias and patient reporting bias is real. There was no objective measurement of headache, and observer bias occurs at the time the observations are recorded. Looking at the charts twice does not alter that. Compliance with binder use was not measured. Unfortunately, the study design limits the strength of the conclusions. Were the patients recognizably similar to yours? The clinical diagnosis of over-shunting was described reasonably well, but headache frequency was only reported in 39 of 70 patients. The binder was used in patients with headaches who did not have typical over-shunting. It is unclear whether these patients were included in the data. There were 14 children whose headaches were mild who were not treated with the binder. Is the therapy feasible in your practice? Yes. The binder and its technique of application were described and illustrated and are likely available to any pediatric neurosurgeon. Were all clinically relevant outcomes reported? No. The paper does not discuss complications related to the binder, except to say that 19 patients were noncompliant with its use. Were clinical importance and statistical significance considered? The reported results seem clinically important. Headache relief was common. Recurrent headache was discussed. Without a control group, there is no basis for comparison and therefore no evaluation of statistical significance. Were all the patients accounted for at the end? Nineteen of the 89 patients did not undergo follow-up (7 were noncompliant, 9 were lost to follow-up, and 3 were admitted to the hospital but the reason for admission is not given). There are additional missing patients in the data on recurrent headache and repeat use of the binder. The authors suggest that “...it is likely that some of the ... patients lost to follow-up ... had favorable responses....” Perhaps, but the opposite is also possible. So, is this a model study for the evaluation of treatment efficacy? No. This is a large experience with a novel low-risk treatment that may help chronic headache sufferers who are often subjected to repeat surgery and its complications. I will keep it in the back of my mind when I encounter these children whose care can be challenging. (http://thejns.org/doi/abs/10.3171/2012.1.PEDS11523)