I read with interest the article by Ng and Sell1 in the May 2004 issue entitled "Predictive value of the duration of sciatica for lumbar discectomy". It is a rigorous assessment of this much debated question, but to call it a prospective, cohort study, despite the fact that the duration of symptoms varies at recruitment, may lead readers to draw a possibly erroneous conclusion.
It is the nature of sciatica from disc prolapse that patients may improve at any time, with a gradual decrease in the incidence of patients experiencing sufficient improvement to obviate the need for surgery. The necessary corollary of this is that the longer a patient has had sciatica, the nearer is he or she to the worse end of the spectrum of disease. This may be the entire explanation of the worse outcome - by waiting longer, one is not predicting a worse result from surgery, but merely selecting a worse group of patients.
What is needed is a cohort study with recruitment based on an intention to treat (by any means - conservative or operative), not on an intention to operate. My feeling is that waiting over a year may indeed be associated with a worse outcome, but unless the dropout rate of patients improving spontaneously during the waiting time is known, it cannot be proven.
M. F. BROWN, PhD, FRCSEd(Orth)
University Hospital of North Staffordshire
1. Ng LCL, Sell P. Predictive value of the duration of sciatica for lumbar discectomy. J Bone Joint Surg [Br] 2004;86-B:546-9.
We thank Mr Brown for his interest in our paper. In answer to his comment, we would consider it unlikely that readers of the Journal would draw an erroneous conclusion on the data presented. The title of the paper makes it clear that the cohort in question is an operative one. Our article contributes to the literature regarding the optimum timing for operation in radicular pain.
The perfect answer to a question often needs to be balanced against the practical realities of clinical practice. The dropout rate to follow-up of a prospective cohort of non-operatively treated patients would be so high that the suggested comparison based on intention to treat could not be achieved in our clinics. A better level of evidence would be a randomised, controlled trial.
Readers of our paper will have noted that patients still improve in terms of pain after a year of symptoms, but that decline in disability as indicated by the outcome measures used showed less change with time, perhaps suggestive of the development of biopsycho-social factors with chronic symptom duration. It is the authors' feeling, that this is the major factor, but it remains to be proven.
P. SELL, BM, MSc, FRCS
L. NG, BM, MRCS
Leicester General Hospital, Leicester, UK.
Copyright British Editorial Society of Bone & Joint Surgery Jun 2005
Provided by ProQuest Information and Learning Company. All rights Reserved