Lee HJ, Chung JW, Chu IT, Kim YC. Comparison of distal chevron osteotomy with and without lateral soft tissue release for the treatment of hallux valgus. Foot Ankle Int. 2010 Apr; 31(4): 291-5.

WHY did the authors undertake this study?
A lateral soft tissue release is commonly performed as an adjuvant procedure for the surgical correction of the hallux abductovalgus deformity. However, some surgeons have questioned its use because of the potential risk for complication. The aim of this study was to “define the necessity” of the lateral soft tissue release in the flexible, mild-moderate hallux abductovalgus deformity.

HOW did they attempt to answer this question?
The primary outcome measures of this study included pre- and post-operative radiographic measurements (1st intermetatarsal angle and the hallux abductus angle), pre- and post-operative 1st MPJ sagittal plane range of motion analysis, and quantification of patient outcomes utilizing the AOFAS scale.

Lateral release was performed through a separate incision and included release of the adductor tendon, lateral capsule, transverse metatarsal ligament and the lateral sesamoid metatarsal ligament. 152 feet in 86 consecutive patients were analyzed in the study, and surgical correction was performed alternatively with and without lateral soft tissue release.

WHAT were the specific results?
No statistically significant differences were found with respect to post-operative radiographic parameters or AOFAS scores. However, there was a statistically significant difference in post-operative 1st MPJ range of motion, with the lateral soft tissue release group demonstrating decreased range of motion (60 degrees versus 76 degrees, p<0.05).

HOW did the authors interpret these results?
From these results, the authors concluded that the lateral soft tissue release may not be associated with improved patient and radiographic outcomes in the flexible hallux abductovalgus deformity, but may be associated with decreased range of motion.

Sevestre MA, Labarere J, Casez P, Bressollette, Haddouche M, Pernod G, Quere I, Bosson JL. Outcomes for inpatients with normal findings on whole-leg ultrasonography: a prospective study. Am J Med. 2010 Feb; 123(2): 158-65. (Pubmed ID#: 20103025)

WHY did the authors undertake this study?
Deep vein thrombosis (DVT) is a common podiatric surgical complication and is most commonly diagnosed with venous ultrasonography. The aim of this study was to analyze outcomes in patients with a suspected DVT, but with a negative diagnostic ultrasound study.

HOW did they attempt to answer this question?
The primary outcome measure of this study was incidence of deep vein thrombosis 3-months following a negative whole-leg ultrasound initially performed to rule-out DVT. Through a prospective multicenter cohort design, nearly 2000 hospital in-patients with suspected DVT were included in the analysis. Twenty percent of patients were found to have a DVT at the initial time of clinical suspicion. The remaining 80% of patients with a negative study were not given anti-coagulant therapy, and were clinically re-evaluated at the 3-month mark.

WHAT were the specific results?
At the 3-month mark following a negative ultrasound, approximately 2% of patients were found to have developed a DVT.

HOW did the authors interpret these results?
From these results, the authors concluded that withholding anticoagulation from patients with a clinical suspicion of DVT but negative ultrasound seemed like safe practice. They caution however that more information is required before recommendations can be made with respect to anti-coagulation and re-scanning.

Let’s take a closer look at the topic of the difference between incidence and prevalence as they apply to these articles. This difference can be challenging to appreciate, but the take-home point is that we typically deal with incidence in the podiatric literature. I like to think of each of these as "rates", or the proportion of a given population that has some variable over some unit of time. The difference between the two is really how we define this unit of time.

The easiest example to appreciate is a complication rate, which is typically an example of an incidence. The Lee et al study had a very familiar study design in our literature: a group of patients underwent a surgical procedure and the authors compared pre- and post-operative variables. One of the post-operative variables that they analyzed was complications, such as recurrence of the deformity. A total of 7 patients in Group A (the group that has the lateral soft tissue release) had recurrence for an incidence of 9%. This incidence rate of 9% represents the proportion (7/74) of a given population (Group A) that had some variable (recurrence of the deformity) over some unit of time (the duration of the follow-up period).

We also saw an incidence rate reported in the Sevestre et al DVT ultrasound study. The incidence rate of 1.9% represents the proportion (10/513) of a given population (patients with clinically suspected DVT, but with a negative initial ultrasound) that had some variable (thromboembolic event) over some unit of time (the 3 months following initial ultrasound).

Prevalence, on the other hand, also represents a proportion of a given population, but over a different unit of time. In fact, the unit of time is essentially an instantaneous “right now”. Take diabetes for example. There were about 25 million diabetics in the United States in 2009. So a prevalence rate of ~8% represents the proportion (~25 million / ~300,000,000 million) of a given population (US citizens) that had some variable (diabetes) over some unit of time (2009). This prevalence rate is the amount of patients that had diabetes right now in 2009, while the incidence rate would be the amount of patients that developed diabetes during 2009.

One can certainly appreciate how incidence and prevalence provide critical readers with different information. Although both describe outcome variables and proportions of the study population, one is more focused on how many people develop some variable over a period of time (incidence), while the other more relates to how many people currently have a given variable (prevalence).