| Vol. 1 Issue 40 |
| PRESENT Journal Club is made possible by a generous grant from:
|  |
|
Sep 1, 2011 |
|
In this issue we examine articles from the following journals The Journal of Foot and Ankle Surgery and the Plastic and Reconstructive Surgery. In addition, in the “critical analysis” section we’ll take a closer look at the specific topic of the situations in which we essentially ignore statistics as it applies to these articles. And finally, please join us for an online discussion of these and other articles on our eTalk page.
|
|
PODIATRY JOURNAL REVIEW |
|
WHY did the authors undertake this study?
Although prophylactic antibiotic administration is an accepted practice with respect to orthopedic surgery, the authors here question whether there is any difference between giving antibiotics prior to tourniquet inflation versus immediately after tourniquet inflation.
HOW did they attempt to answer this question?
The primary outcome measures of the study were the development of a post-operative wound infection, interval to wound healing and subjective patient outcome (“satisfied” versus “unsatisfied”).
One hundred-six patients were prospectively enrolled and randomized to receive either prophylactic antibiotic administration 5 minutes before exsanguination and inflation of tourniquet or 1 minute after exsanguination and inflation of tourniquet. All patients underwent clean, elective orthopedic procedures (an unusual mix of tibial fractures, ankle fractures, hallux valgus and clubfoot).
WHAT were the specific results?
Statistically significant differences were found with respect to all primary outcome measures. Patients receiving antibiotics after tourniquet inflation were less likely to have a post-operative wound infection (3.9% vs. 14.8%; p=0.031), more likely to heal faster (3.0 weeks vs. 4.0 weeks; p=0.002), and were more likely to be subjectively satisfied with the procedure (100% vs. 85.2%; p=0.005).
HOW did the authors interpret these results?
Based on these results, the authors concluded that administration of prophylactic antibiotics prior to tourniquet inflation may not give better results when compared to administration prior to tourniquet inflation.
There are several other articles in this issue that readers may find both beneficial and interesting. Schwatzkopf et al investigate foot and shoe size mismatch, one of my clinical pet peeves. Roukis examines publication rates of manuscripts presented at the ACFAS conference, while Jordan et al discuss the radiographic fate of the syndesmosis after trans-syndesmotic screw removal, a manuscript presented at ACFAS! The Cook’s lead a task force to examine the validity and reliability of the ACFAS scoring scales. Imade et al go to a lot of trouble to fix a hallux proximal phalanx non-union. And Ross, Catanzariti, and Mendicino provide an instructional guide for the hematoma block prior to ankle relocation.
|
|
|
MEDICAL JOURNAL REVIEW |
|
WHY did the authors undertake this study?
The angiosome concept of arterial anatomy was developed by GI Taylor and refined for the lower extremity by Attinger. The concept describes a three dimensional block of tissue (bone, muscle, fascia, skin) supplied by a single source artery. A newer concept is that of the perforasome, or a region of tissue (primarily skin and superficial fascia) supplied by a single cutaneous perforating artery. This information can be valuable with respect to planning and design of flaps. The authors of this study aimed to describe the cutaneous perforators of the lateral aspect of the lower leg.
HOW did they attempt to answer this question?
The lateral compartment of fifty-two cadaveric limbs were dissected to identify the presence and course of any cutaneous artery that penetrated the deep fascia with a diameter greater than 0.5mm. Methylene blue was utilized in 10 of the limbs.
WHAT were the specific results?
Cutaneous perforators primarily came from the tibioperoneal trunk proximally, the peroneal artery centrally and the posterior tibial artery distally, with the majority of perforators originating from the central zone. They were primarily septocutaneous, meaning that the cutaneous perforator branched from the underlying source artery and penetrated the deep fascia at the level of an intramuscular septum.
HOW did the authors interpret these results?
Based on these results, the authors concluded that knowledge of the leg perforasome concept can be reliably utilized for the planning of perforator-based flaps harvested from this region.
There are several other articles in this and other issues that readers may find both beneficial and interesting. Ducic and Attinger the role of diabetes in pedicled muscle flaps and free flaps. They find that although diabetic patients undergoing these procedures tend to have longer post-operative courses, overall patient survival is encouraging compared to historical controls with major amputation. Veber et al quantify the arc of rotation of medial gastrocnemius flaps with several different dissection techniques. Davis and Drew propose the use of standardized patients as part of a residency training curriculum. And two free articles, ( article 1, article 2 ) take an evidence-based look at the levels of clinical evidence with respect to plastic surgery.
|
|
|
CRITICAL ANALYSIS OF THE LITERATURE |
|
Let’s take a closer look at the topic of the situations in which we essentially ignore statistics, particularly as it applies to the Akinyoola et al tourniquet antibiotic article. As anyone who has ever been in my office can attest, I have several favorite quotes about statistics:
- Someone said: A man with one leg in boiling water and the other in ice is, on average, comfortable.
- Easterbrook said: Torture numbers and they’ll confess to anything.
- Someone said: Statistics can be made to prove anything, even the truth.
- McHale said: The average human has one testicle.
The obvious point is that statistics never have and never will prove anything; they simply give us a degree of confidence with which we can judge our results and/or findings. As you certainly know, we usually use 0.05 or 5% as the “cut-off” for the level of statistical significance in medical research. This is a bit overly simplified, but this implies that if we found a statistically significant difference between two groups of numbers, then we would probably find a similar statistical significance 94 more times if we re-ran the investigation 100 total times. But it also means that we wouldn’t find a statistically significant difference 5 out of those 100 times. In other words, we’re probably right, but we’re not 100% confident that we’re right, and we have about a 5% chance of being wrong.
The Akinyoola et al article examined one cohort of patients: individuals who underwent an elective lower extremity orthopedic procedure with the use of tourniquet hemostasis. The patients were prospectively randomized to receive either (1) prophylactic intravenous antibiotics 5 minutes before exsanguination and tourniquet inflation or (2) prophylactic intravenous antibiotics 1 minute after exsanguination and tourniquet inflation. As one of their primary outcome measures, the authors found a post-operative wound infection rate of 14.8% in the before group and of 3.9% in the after group, and they found this difference to be statistically significant (p=0.031).
Stop the presses! I now have Level II evidence in support of prophylactic antibiotic administration after exanguination and tourniquet inflation. Should I just ignore all those JCAHO posters around my operating room urging me to deliver antibiotics within 1 hour prior to incision?!? The answer is of course not. Our authors here did not even conclude that antibiotic administration after tourniquet inflation was superior, just that it “did not give better results”, even calling their findings “ironic”.
It’s funny that when the statistics tell us something that goes against our hypothesis, or our expectations, or our common sense, then we are quick to ascribe the findings to chance, or to that 5% chance of being wrong that we expect. We’re not so quick to do the same thing in the opposite situation, or question that our results may be wrong when they support our expectations.
An expected 5% chance of being wrong certainly doesn’t seem very significant, but it’s really about 1 out of 20 times when you think about it. Once upon a time I did a nerdy study examining JFAS articles and found that they published 172 articles over a 5 year period that utilized comparative statistics, and most using multiple comparative statistical techniques. The numbers tell us that, just by chance, about 9 of those articles found a statistically significant difference when there wasn’t one, with the authors drawing conclusions accordingly. This is essentially just chance, and doesn’t include all the methodological problems with study design that may unduly influence results and form the bulk of what we usually expound upon in this space. Let’s listen to Watt who said, “Do not put your faith in statistics until you have carefully considered what they do not say”. |
|
DISCUSSION
|
Please join us for an online discussion of these topics: |
|
|
|
I hope you find PRESENT Journal Club a valuable resource. Look out for the eZine in your inbox. Please do not hesitate to contact me if there is anything I can do to make this a more educational and clinically relevant journal club. |
Andrew Meyr, DPM
PRESENT Podiatry Journal Club Editor
Assistant Professor, Department of Podiatric Surgery,
Temple University School of Podiatric Medicine,
Philadelphia, Pennsylvania
[email protected]
Get a steady stream of all the NEW PRESENT Podiatry
eLearning by becoming our Facebook Fan.
Effective eLearning and a Colleague Network await you. |
 |
|
