Drops from a needle. The image is CC by Evan Leeson
Many of us orthopaedic surgeons have been frustrated by waiting for the anesthesiologist to finish with the spinal anesthesia. It is therefore of great relief that Pugely et al. write that this frustration is not in vain. Patients that receive spinal anesthesia seem to have fewer complications after a total knee arthroplasty than those with general anesthesia. A conclusion based on a large registry study with 14 000 patients although the overall odds ratio was though not that alarming, 1.3.
Although it is a nice registry study with well performed statistics, I lack a few details:
- What is the coverage and completeness of this amazingly detailed database (American College of Surgeons National Surgical Quality Improvement Program) that they’ve used?
- Has the database been validated by external reviews?
- Why couldn’t they include the hospital volume by just calculating the number of cases at that hospital? They had access to the Medicare database, all you need is a simple script…
- As age was the strongest variable, wouldn’t it have been better to model it as a continuous variable? The small differences in complication rates could be just residual confounding due to cut point bias.
- An imputated result should have been reported as well – 10 % of the patients were excluded due to missing data (not including the above completeness issue). This calls for some kind of missing data analysis.
Apart from these study issues, I’m also a little frustrated that not more effort was put into graphics. A nice forest plot would perhaps make the regressions more visually appealing.