Classifications in Brief: Lawrence and Botte Classification of Fifth Metatarsal Fractures

History

Fractures of the fifth metatarsal base are common, and their clinical management varies depending on the fracture location. Blood supply to the fifth metatarsal varies, with some areas receiving limited vascularization, particularly in the 4-5 intermetatarsal region [14-16]. Consequently, treatment approaches and healing times are influenced by the specific anatomic site of the fracture [4, 8, 13]. Sir Robert Jones first described a metaphyseal-diaphyseal fracture at the base of the fifth metatarsal in 1902, and the injury since then has been termed the “Jones fracture” [5]. Today, the term “Jones fracture” is sometimes used erroneously to describe all fifth metatarsal base fractures; however, a Jones fracture specifically refers to fractures in the specific region at the metaphyseal-diaphyseal junction. In 1960, Stewart [17], understanding the variability of the behavior of these fractures, divided these injuries into five types according to their locations and morphologies. Later, Torg et al. [19] published a classification based on the apparent amount of time since injury, according to the degree of sclerosis found on imaging. Lawrence and Botte [6], recognizing the controversy and lack of universal guidelines in the treatment of these injuries, introduced their classification in 1993. Their classification assigned fractures into three anatomic zones of the proximal fifth metatarsal for purposes of guiding treatment.

Purpose

The Lawrence and Botte classification [6, 8] was initially written to establish a treatment algorithm based on both the location of the fracture and the desires of the patient. The goal of the classification scheme is to provide a more standardized and reproducible system to help guide treatment, with each zone and subtype of injury having a respective treatment recommendation. However, low levels of intraobserver and interobserver reliability that have been observed in several studies evaluating this classification [3, 9, 11] suggest that the classification does not allow surgeons and clinician scientists to achieve those goals, and as such, this classification should not be used for those purposes.

Description

The Lawrence and Botte classification categorizes fractures of the proximal fifth metatarsal into three distinct zones (Fig. 1). These zones are delineated based on their anatomic location and are intended to be assigned by clinicians when interpreting plain radiographs (Fig. 2). Zone 1 injuries, referred to as tuberosity avulsion fractures, represent 93% of all injuries according to Bowes and Buckley [1]. Zone 2 Jones fractures of the metaphyseal-diaphyseal region account for 4%, and Zone 3 diaphyseal stress fractures account for the remaining 3% [1, 18]. Zone 3 fractures are further classified into subtypes according to Torg’s earlier classification of diaphyseal stress fractures based on the degree of sclerosis [19]. Type I fractures are acute, presenting with a narrow fracture line without fibrosis. Type II (delayed union) is characterized by a widened fracture line with intramedullary sclerosis, indicating ongoing bone stress. Type III refers to symptomatic nonunions, characterized by the complete obliteration of the medullary canal attributed to sclerotic bone formation [6, 19].

Fig. 1.

In the Lawrence and Botte classification, injuries are divided into one of three zones according to anatomic location.

Fig. 2.

These radiographs show the Lawrence and Botte classification of fifth metatarsal fractures: (A) Zone 1, tuberosity avulsion fracture; (B) Zone 2, metaphyseal-diaphyseal “Jones fracture”; and (C) Zone 3, diaphyseal stress fracture. A color image accompanies the online version of this article.

Zone 1 fractures, commonly known as tuberosity avulsion fractures and which occur near the insertion of the peroneus brevis tendon, are treated symptomatically and usually nonoperatively [6]. For Zone 1 fractures with displacement > 3 mm in any plane, operative treatment such as open reduction and internal fixation or closed reduction and pinning may be considered [1].

Zone 2 Jones fractures that are nondisplaced can be treated with nonweightbearing immobilization for 6 to 8 weeks. In athletes and active patients, operative treatment may be considered for Zone 2 Jones fractures, particularly in cases where early return to activity is of priority. Numerous studies in athletes have demonstrated lower risk of nonunion and subsequent refractures in patients treated surgically with intramedullary screw fixation compared with nonoperative treatment [10, 12].

Zone 3 diaphyseal stress fractures are further subcategorized as acute (I), delayed union (II), and symptomatic nonunion (III): Zone 3 Type I fractures are treated nonoperatively, and the authors recommend nonweightbearing and immobilization for up to 20 weeks [6]. Zone 3 Type II injuries are commonly treated with either nonweightbearing immobilization or operative intervention, either with an intramedullary screw or compression plating, depending on a shared decision made by patient and surgeon. Use of bone graft and postoperative rehabilitation remains controversial and often is surgeon dependent. Zone 3 Type III symptomatic nonunions usually are treated surgically [6].

The prognosis of fifth metatarsal base fractures is described in the original paper as “excellent healing potential,” “good healing potential,” and “variable healing potential” for Zones 1, 2, and 3, respectively [6]. Zone 1 fractures are associated with excellent healing potential because of the adequate vascularity of the region [16]. Furthermore, these fractures are predominantly nondisplaced, which further indicates nonoperative treatment modalities. Conversely, Zone 3 and Zone 2 Jones fractures are associated with higher anatomic stress and have decreased vascular supply [6, 14-16], which suggests that nonweightbearing treatment should be employed, or in some patients, that surgery should be considered either acutely or to address a symptomatic nonunion [6].

Validation

There are three studies we are aware of that have assessed the interrater and intrarater reliability of the Lawrence and Botte classification [3, 9, 11], and none documented a level of interrater reliability that was sufficiently high for us to recommend the widespread use of this classification schema.

The referenced studies assessed the agreement of participants viewing radiographs using a kappa (κ) coefficient instead of observed agreement. The κ coefficient analyzes the level of agreement between observers that is beyond the agreement that would occur because of chance alone. κ coefficient values range from 0 to 1, with a value of 0 representing agreement that occurs completely by chance and 1 representing complete agreement [7]. κ coefficient values of 0.40 to 0.59 are considered weak, 0.60 to 0.79 are considered moderate, 0.80 to 0.90 are considered strong, and > 0.90 are considered almost perfect levels of agreement. Intrarater agreement involves the same observer and assesses the level of repeated agreement between time points. Interrater agreement assesses agreement between different observers of the same patients [7].

Michalski et al. [9] and Noori et al. [11] evaluated fellowship-trained orthopaedic surgeons and their interrater agreement of the Lawrence and Botte classification system, with κ = 0.537 [9] and κ = 0.66 [11] indicating weak and moderate interrater agreement, respectively. Chapman et al. [3] evaluated two medical students trained in the classification system without prior knowledge of it; they also reported moderate interrater agreement κ values for all zones.

In contrast, two of these studies also assessed intrarater reliability, and both demonstrated higher κ values. Here, Michalski et al. [9] reported that the intrarater reliability ranged from κ = 0.60 to κ = 0.90, and Noori et al. [11] reported a median intrarater reliability of κ = 0.75.

All three existing studies found that the Lawrence and Botte classification demonstrated interrater reliability that may be insufficient for consistent use in research or clinical decision-making [3, 9, 11]. This suggests that modification to the existing Lawrence and Botte classification should be considered, or that a new system altogether that delivers greater reliability should be developed. Noori et al. [11] proposed this in their study in which participants used a modified version of the Lawrence and Botte classification, wherein Zones 2 and 3 were combined into a single Zone 2. When interrater agreement was assessed using the modified classification, the κ value improved from 0.66 using the classic three-part Lawrence and Botte classification to 0.83 with the simple two-part classification [11]. This was also observed by Michalski et al. [9] when using a two-part metaphyseal/metadiaphyseal system; the κ value improved from 0.537 to 0.705 for interrater reliability when using a simplified two-zone system. It is important to acknowledge that the observed improvement in κ values may be partly attributed to the reduction in classification categories rather than an inherently superior classification system. Future studies should evaluate whether this simplification offers meaningful clinical advantages beyond increased statistical agreement.

Despite interrater reliability ranging from weak to moderate [3, 9, 11], the Lawrence and Botte classification continues to be widely used in both research and clinical practice [2, 11]. Its continued use in research and clinical settings could be problematic, as it may lead to inconsistent, variable, or misleading results.

Limitations

The Lawrence and Botte classification system has several limitations, with the most notable being its weak to moderate interrater reliability [3, 9, 11]. Some of the authors of these validation studies recommend that clinicians consider using an alternative system with higher consistency and reliability.

The difficulty in distinguishing between Zone 2 and Zone 3 fractures is further compounded by unclear differences in clinical outcomes, including patient-reported satisfaction, radiographic nonunion risk, and functional recovery, such as return to sport [4]. A retrospective study by Chuckpaiwong et al. [4], which assessed 61 proximal fifth metatarsal fractures classified as Zone 2 or Zone 3, found that patients with injuries in those zones were treated similarly and achieved similar radiographic union, return-to-sport times, and patient satisfaction. This has been used as an argument for simplifying the Lawrence and Botte classification from a three-part to a two-part system. However, it is important to note that this study did not account for potential differences in fracture etiology, such as chronic lateral overload from cavovarus alignment, which may be more prevalent in Zone 3 fractures. Additionally, as a retrospective study, treatment decisions were influenced by the attending surgeon, who may have considered factors such as foot alignment when selecting a management approach, which may create a selection bias. These limitations suggest that while the study provides valuable insight into outcomes, further prospective research is needed before concluding that merging Zones 2 and 3 is clinically beneficial.

Conclusion

The Lawrence and Botte classification was created as a tool to guide the treatment of proximal fifth metatarsal fractures based on their anatomic location [6]. Although seemingly straightforward, its use in clinical and research settings has faced challenges because of its limited interrater reliability [3, 9, 11]. Forementioned studies have suggested combining Zones 2 and 3 into a single zone [9, 11], as removing this distinction may improve the interrater reliability [9, 11]. However, current treatments for Zones 2 and 3 are different; thus, combining these zones would alter how the classification system guides treatment. While some research suggests that differentiating between Zones 2 and 3 injuries may not be necessary [4], further investigation is needed to confirm this.

Many orthopaedic injuries lack a useful and valid classification system, and fractures of the fifth metatarsal are likely one of them. While the Lawrence and Botte classification can sometimes aid in categorization, its limitations in guiding treatment decisions and ensuring consistent communication raise questions about its effectiveness. Further research is needed to determine whether refining the existing system or developing a novel, more reliable classification would enhance reproducibility and improve clinical outcomes, including functional recovery and patient satisfaction, before clinical implementation.