Article for review: Hamid N, Loeffler BJ, Braddy W, Kellam JF, Cohen BE, Bosse MJ.  Outcome after fixation of ankle fractures with an injury to the syndesmosis: the effect of the syndesmosis screw.  J Bone Joint Surg Br.  2009 Aug; 91(8): 1069-73. (PubMed ID: 19651836). Follow this link.

WHY did the authors undertake this study?
The authors of this study aimed to compare clinical and radiographic outcomes of patients with intact, broken and removed syndesmotic screws following open reduction and internal fixation of ankle fractures involving syndesmotic disruption.  The general topic of syndesmotic fixation remains one of great debate within the podiatric and orthopedic communities, with specific questions ranging from whether one or multiple screws should be used, which type of screws should be used,  the size of these screws, how many cortices should be engaged, where these screws should be placed, foot position during screw placement, etc.  Although freely acknowledging that all of these questions will not be answered with this study, the authors hoped to provide evidence towards the specific question of whether or not intact or broken screws should be removed following syndesmosis healing.

HOW did they attempt to answer this question?
The authors collected data on a total of six outcome measures in this study:  (1) whether the syndesmotic screw was removed, remained retained/intact, or remained retained/broken; (2) whether there was radiolucency around retained screws; (3) whether there was site tenderness around retained screws; (4) AOFAS ankle/hindfoot post-operative scores; (5) visual analogue scale post-operative scores; and (6) a measurement of post-operative tibio-fibular clear space.  

Inclusion criteria of the population cohort were patients who underwent open reduction and internal fixation of an ankle fracture with screw stabilization of a disrupted syndesmosis, and who followed up for at least one year for clinical and radiographic evaluation.  Patients were excluded if they were less than 18 years of age, suffered from chronic syndesmotic complaints, had bioabsorbable syndesmotic screw placement, or had one of several post-operative complications.  A total of five surgeons performed all of the surgical procedures, two of whom routinely remove syndesmotic screws at the 12-week mark.

WHAT were the specific results? 
When examining all of the outcome measures, only one statistically significant difference was reported between the three groups of patients (removed syndesmotic screw, retained/intact syndesmotic screw, retained/broken syndesmotic screw).  Post-operative AOFAS scores in the retained/broken group had the best outcome (p=0.0466), while there was no statistically significant difference between the removed syndesmotic screw and retained/intact syndesmotic screw groups.

HOW did the authors interpret these results?
From these results, the authors concluded that there was no difference in clinical and radiographic outcomes of patients with intact or removed syndesmotic screws.  They also concluded that their data did not support the removal of intact or broken syndesmotic screws.

Article for review:  Weigelt C, Rose B, Poschen U, Ziegler D, Friese G, Kempf K, Koenig W, Martin S, Herder C.  Immune mediators in patients with acute diabetic foot syndrome.  Diabetes Care.  2009 Aug; 32 (8): 1491-6. (PubMed ID#: 19509015). Follow this link.

WHY did the authors undertake this study?
Diabetes has been called a “disease of complications”, and nearly all of these complications (MI, stroke, nephropathy, neuropathy, vasculopathy, etc) can be related back to the systemic inflammation and low-grade immune activation seen in diabetic patients.  The aim of this study was to compare systemic immune mediators in diabetic patients with and without lower extremity ulceration to identify if diabetic foot disease is associated with an underlying systemic immune activation. 

HOW did they attempt to answer this question?
The primary outcome measures in this study were circulating levels of a wide variety of acute-phase proteins, cytokines and chemokines known to be associated with inflammation and immune function. 

Inclusion criteria of the study population were consecutive diabetic patients seen at the authors’ Diabetes Clinic in Germany.  Patients were then stratified into two groups:  those without a history of foot ulcer (n=140) and those with an acute foot ulcer (n=170).  Data was collected at the time of patient presentation

WHAT were the specific results?
The results demonstrated specific and nonrandom alterations in immune activation between the two groups.  Some factors (CRP, fibrinogen, IL-6, etc) were significantly upregulated in patients with foot ulceration, while some factors (IL-8, IL-18 and IP-10) were unaffected and others still (RANTES) were actually significantly downregulated.  These statistically significant results were associated with severity of the ulceration, but independent of the presence of infection and other potentially confounding patient demographics/comorbidities.

HOW did the authors interpret these results?
From this data the authors concluded that the presence and severity of lower extremity ulceration was significantly associated with the upregulation of certain acute-phase inflammatory factors within specific immune pathways, as opposed to a general systemic immune activation.  This data may serve as a starting point for directed anti-inflammatory intervention for diabetic foot ulceration treatment and prevention.

With a basic understanding of these two questions, one can account for over 95% of the statistical techniques utilized in the podiatric medical literature.  Over the next several eZines, we’ll begin to break down these two questions and start to build a foundation of statistical knowledge.

The Hamid et al article actually provides a wealth of teaching/talking points with respect to critical analysis of the medical literature.  Hopefully we can get to a lot of these in our eTalk, but the one that I would like to focus on now is the data type of the outcome measures.  This gets into our first question of:  What type of data is being analyzed?  The most basic way to classify data is to determine whether it is categorical or quantitative, and the Hamid article provides good examples of both.

Categorical data (also known as nominal data) simply implies that you can group data into different categories, but those categories have no defined quantity or arithmetic relationship.  The Hamid article had three outcome measures that fit this description:  (1) whether the syndesmotic screw was removed, remained retained/intact, or remained retained/broken, (2) whether there was radiolucency around retained screws, and (3) whether there was site tenderness around retained screws.  So either patients had radiolucency around retained screws or they didn’t.  It’s either one or the other, one category isn’t more or less than the other, and there is no middle ground.  Other common examples of categorical data that we deal with every day include gender, race, and whether or not a patient has some co-morbidity.

Quantitative data, on the other hand, implies that you can assign a number to the data.  The other three outcome measures from the Hamid article fit this description:  (4) AOFAS ankle/hindfoot post-operative scores, (5) visual analogue scale post-operative scores, and (6) a measurement of post-operative tibio-fibular clear space.  So a patient can have a visual analogue score anywhere between 0 and 10, and a visual analog score of 7 is more or larger than a visual analog score of 3.  With quantitative data, there is an arithmetic relationship between two different categories.   Other common examples of quantitative data that we deal with every day include age, radiographic angles, income, etc.

Quantitative data can be further broken down, but we'll get to that in a subsequent eZine. Just as the take-home point from the first eZine was identification of outcome measures, the point of emphasis from this session is to determine whether the outcome measures are categorical or quantitative. 

Please join us for an online discussion of these articles:

-What have your experiences been with syndesmotic screw fixation? 
 Any strong opinions?
-Does anyone remove all internal fixation as a rule?
-Can any residents give me other good examples or categorical
  and quantitative data?