The reliability of a classification system requires comparison to the gold standard [7]. Furthermore, the validity of a classification system is dependent upon the accuracy with which the system describes the true pathologic process. Audige’s quality criteria from 2004 reflected the importance of clearly described categories and inclusion/exclusion criteria for determining inter- and intra-observer reliability [7].

According to Garbuz et al., a classification system should help orthopaedic surgeons characterize a problem, suggest a potential prognosis, offer guidance in determining optimal treatment, characterize the nature of a problem, and influence treatment decision-making, ultimately improving outcomes [8]. The same authors further asserted that a classification system should form a basis for uniform reporting of treatments [8].

Stratifying patients with stress fractures into prognostic and treatment groups has historically been difficult given the lack of a single widely applicable standard classification system. Textbooks and review articles have cited techniques for describing stress fractures at a particular location, but have rarely been validated as a method for determination of stress fracture severity, risk, and prognosis [9–16]. An understanding of the basic science of fatigue failure of bone correlates well with our clinical experience that structural failure occurs along a spectrum from micro-fractures to complete structural failure.

Because stress fractures have various degrees of structural failure and healing potential, it is important that we develop standardized categorization and descriptive instruments. Descriptive systems should identify the clinically relevant attributes of the injury in a reproducible fashion and should do so in a simple, inexpensive, safe, and widely applicable manner. For a comprehensive description of stress fractures, these characteristics should be incorporated into a system that describes not only the extent of the structural damage but also the healing potential.

Unlike most traumatic fractures, in the case of stress injuries of bone the size and extent of the fracture line vary greatly, and the healing potential varies by location. Some locations typically heal very readily. Other locations, such as the junction of the metaphysis and the diaphysis of the proximal fifth metatarsal, tend to have an increased risk of delayed union, nonunion, and refracture. Boden et al. described high-risk and low-risk stress fractures by their location [17, 18]. Those locations that have a tendency toward delayed union, nonunion, or refracture are classified as high-risk stress fractures. The varied healing potentials may be related to biologic and/or biomechanical factors of the different anatomic sites.

An important distinction regarding stress fractures is whether they are high- or low-risk fractures (Table 4.1). This classification system has been proposed many times in the literature [17–22]. Such a system provides a reproducible way for medical personnel to determine the course of treatment and the timeframe of recovery before the athlete can return to play. Stress fractures are considered to be high-risk fractures if they have any of the following characteristics. First, these fractures have a predilection to progress to complete fracture (fifth metatarsal), delayed union (anterior cortex tibia), or nonunion (tarsal navicular). Second, a delay in diagnosis and treatment can either prolong an athlete’s non-weightbearing status and his or her restriction from sport, or change a nonsurgical treatment to one requiring operative fixation with or without bone graft.

These high-risk sites possess a common biomechanical characteristic [3–5, 17, 18]. The initiation of their associated fracture lines typically occurs on the tension side of the bone or in a watershed (relatively avascular) area of the vascular supply (e.g., superior side of the femoral neck, anterior cortex of the tibial shaft, lateral aspect of the proximal fifth metatarsal, and the dorsal side of the tarsal navicular) [5]. Because bone is less resistant to tensile than compressive forces, this likely puts the bone at these locations at increased risk for microcrack initiation. Why these “high-risk” locations have an increased risk of impaired healing is likely a result of additional influences beyond the biomechanical factors.

The biomechanical factors of being on the tensile side of the bone explain the increased requirement for a healing response, but biologic factors may come into play as well. For example, the proximal junction of the fifth metatarsal diaphysis/metaphysis is a vascular watershed area with suboptimal blood supply to support fracture healing. Locations of high-risk stress fractures may be the combination of increased micro-failure due to biomechanical conditions coupled with impaired biologic healing capacity. Table 4.1 lists the locations of commonly described high-risk stress fractures.

A common example of a poor natural history of a high-risk stress fracture is neglect of an early proximal fifth metatarsal stress failure that results in either an acute fracture or, should it heal, a subsequent refracture. Recognition of this fracture as a high-risk location and early intramedullary screw fixation will often lead to timely healing, and the athlete can resume his or her career with a markedly decreased risk of re-injury.

When compared with high-risk fractures, low-risk fractures have an overall favorable natural history. In contrast to high-risk fractures, which tend to be on the tension side of bone, low-risk fractures tend to occur on the compression side of bone and typically heal readily. Low-risk fractures are less likely to develop a delayed or nonunion, recur, or have a significant complication should it progress to complete fracture. Low-risk stress fractures can typically be treated with activity modification and rarely require surgical intervention. Low-risk fractures include the femoral shaft, medial tibia, ribs, ulna shaft, and first through fourth metatarsals. Anatomic location of the fatigue bone failure is the distinguishing characteristic between high- and low-risk fractures. Determining whether the stress fracture is in a high-risk vs. a low-risk location is key to optimal care as it impacts both treatment and prognosis discussions. This characteristic makes judging a stress fracture to be either high-risk or low-risk an important element in the “classification” of the injury. Table 4.2 describes the key elements of high-risk vs. low-risk stress fractures.

The goal in treating athletes is to make an expeditious diagnosis of a stress fracture because those with stress fractures classified as low-risk fractures can participate in modified sports activity, whereas athletes with stress fractures classified as high-risk should be aggressively managed with non-weightbearing activity or surgery [3–5, 19, 20]. This obviously important clinical implication of the fracture being identified as either high- or low-risk makes it one of the most important classifications of fatigue failure of bone the clinician can make.

A recent literature review by Miller et al. revealed 26 stress fracture classification systems [21]. Table 4.3 lists the classification systems reviewed [17, 18, 23–44]. The goal of this review was to determine what classification and grading systems have been referenced in the literature for stress fractures. At the outset of this review, the authors of this study asked two questions: (1) “What classification systems are used in the evaluation and treatment of stress fractures?” and (2) “What are the features of each classification system?” It is clear from their review that many classification systems have been developed and applied to stress fractures since Breithaupt first categorized the injury in 1855 [1]. In 42 articles and citations, 27 classification systems were described or referenced.