Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Standard discectomy has better short-term outcomes than conservative treatment but long-term outcomes are similar

ACP J Club. 1994 Jan-Feb;120:15. doi:10.7326/ACPJC-1994-120-1-015

Source Citation

Hoffman RM, Wheeler KJ, Deyo RA. Surgery for herniated lumbar discs: a literature synthesis. J Gen Intern Med. 1993 Sep;8:487-96.



To evaluate the risks and benefits of surgery for herniated lumbar discs.

Data sources

English-language studies were identified using MEDLINE (1966 through 1991) with the following MeSH headings and key words: lumbar, backache, herniated disc, intervertebral disc displacement, sciatica, laminectomy, disc surgery, discectomy, microdiscectomy, and percutaneous discectomy. Reference lists of retrieved articles, book bibliographies, and experts' files were reviewed for additional studies.

Study selection

Studies were selected if the sample size was ≥ 30, if the mean age of patients was ≥ 30 years, if baseline and follow-up data were available for ≥ 75% of all patients receiving surgery, and if the minimum follow-up was ≥ 1 year or if the mean follow-up was ≥ 24 months. Review articles, studies with ≥ 10% of patients with a primary diagnosis other than herniated disc, and studies of anterior discectomy or spinal fusion were excluded.

Data extraction

Data pertaining to study design, patient demographics, clinical description, diagnostic testing, surgical techniques, operative findings, clinical outcomes, reoperations, and operative complications were extracted. A successful outcome was defined as the sciatica being completely absent or occasionally mild with minimal or no restriction of physical activity and a return to previous employment. Point estimates and 95% confidence intervals of the proportion of successful outcomes, complications, and reoperations were obtained by fitting a random-effects logistic regression model.

Main results

81 studies met the inclusion criteria. Most studies had substantial design flaws; only 4 were prospective comparison studies. The 2 randomized controlled trials of standard discectomy showed better 1-year patient- or physician-rated outcomes after surgery (65% to 85%) compared with either conservative treatment (36%) or chymopapain (63%), but longer-term outcomes did not differ. Data from randomized trials were not available for microdiscectomy or percutaneous discectomy. Approximately 10% of patients who received discectomy had further back surgery, and this rate increased over time. Reoperations occurred most frequently after percutaneous discectomy. Serious complications after discectomy were rare. The mortality rate was < 0.1%, the rates of deep wound infection and permanent nerve root damage were < 1%, and the rates of thromboembolic events, wound infections, and discitis were < 2%.


Standard discectomy seems to provide better short-term outcomes than conservative treatment, but long-term outcomes are similar. Discectomies are safe, but reoperations are common and increase over time.

Sources of funding: Seattle Veterans Affairs Medical Center and Agency for Health Care Policy and Research.

For article reprint: Dr. R.M. Hoffman, Arizona Health Sciences Center, 1501 North Campbell Avenue, Room 6335, Tucson, AZ 85724, USA. FAX 602-575-8343.


The review by Hoffman and colleagues is another assault on the literature of backache. Again, we learn that the literature is voluminous but is mostly inadequate for quantitative review. The authors have therefore drawn their inferences from a small subset of studies. The results reiterate the precedent literature. In their dispassionate service to objectivity, the authors fail to stress the salient messages of their exercise. Here is what is left unsaid.

First, herniated lumbar discs never require surgery. A small percentage of patients with regional radicular pain benefit from surgery, often done to alter discal structure. Herniated lumbar discs are common and usually asymptomatic.

Second, there is no compelling evidence that surgery for a regional backache benefits any patients. There is the suggestion of benefit for subacute sciatica but not for pain in the low back.

Third, there is compelling evidence that whatever the indication (generally sciatica), patients are less well served by chemonucleolysis than traditional laminectomy. Similarly, they are less well served by percutaneous microdiscectomy than by chemonucleolysis. Each year tens of thousands of persons with backache are misled into considering these procedures as "advances."

Finally, the benefit afforded the patient with subacute sciatica is not uniform across series. It varies in both relative and absolute magnitude of likelihood. In the absence of progressive neurologic compromise, laminectomy is an elective procedure even for subacute sciatica.

This literature synthesis spans the 3 decades during which the Federal Drug Administration (FDA) was empowered by the Kefauver-Harris amendments to demand a clearly favorable benefit-to-risk ratio before any pharmaceutical could be purveyed. If those amendments had added surgery for regional back-pain syndromes to the purview of the FDA, many American backs would have been spared the scar that denotes an unproved remedy. Judging from the incontrovertible message of the literature, this empowerment is long overdue.

Nortin M. Hadler, MD
University of North CarolinaChapel Hill, North Carolina, USA