Navicular Stress Fracture

The navicular is the most common tarsal bone to sustain a stress fracture, and navicular stress fractures represent approximately 15-35 percent of all stress fractures in athletes. It is an insidious, high-risk, slow-healing injury, and because the initial X-ray is normal in the majority of cases the diagnosis is missed in up to 80 percent of patients at first presentation. The typical delay from symptom onset is 4-7 months, and that delay is the most common cause of a poor outcome.

Who. The highest incidence is in track and field: sprinters, hurdlers and middle-distance runners. Jumpers (basketball, netball, ballet, gymnastics), cutting-sport athletes (football, soccer, rugby) and military recruits under the high volume of repetitive impact loading of basic training follow. In a large military cohort, navicular stress fractures accounted for the longest average time to return to duty of any lower-extremity stress fracture.

What it costs. Return to sport takes 4-6 months even with optimal care. Career impact is documented in elite athletes, and the injury may end a professional career if mismanaged.

The high-risk family. The aggressive treatment of the navicular only makes sense within the high-risk versus low-risk stress-fracture framework. High-risk stress fractures occur on the tension side or in a watershed, poorly vascularised zone, so they tend to progress, displace, go to nonunion or AVN, and need aggressive treatment: strict non-weight-bearing and/or internal fixation, with imaging-confirmed union before return. Low-risk stress fractures are compression-side and well-vascularised, and usually heal with relative rest and activity modification.

• High-risk: tarsal navicular, femoral neck (tension/superior side), anterior tibial cortex ("dreaded black line"), medial malleolus, proximal fifth metatarsal (Jones, zone 2), great-toe (hallux) sesamoid, talar neck/body
• Low-risk: femoral-neck compression side, posteromedial tibial shaft, metatarsal shafts 2-4, calcaneus, distal fibula

The navicular is one of the four classic high-risk stress fractures, alongside the femoral neck tension side, the anterior tibial cortex and the fifth metatarsal metaphyseal-diaphyseal junction, and it sits there because of its central-third watershed. Risk here means a propensity to delayed union, nonunion or complete fracture when missed, which is exactly why a "minor" incomplete dorsal-cortex fracture is treated like a true fracture, with the high-risk protocol from the outset.

Scope. This page covers the stress fracture of the tarsal navicular. Acute traumatic navicular fractures (high-energy body fractures, tuberosity avulsions, the Sangeorzan classification and their operative management) are a different entity covered on the Navicular Fractures page. The two share the bone and the watershed anatomy but differ in mechanism, imaging threshold and treatment.

View larger

The watershed. The central third of the navicular is a vascular watershed. The medial side is perfused by branches of the posterior tibial artery and the lateral side by branches of the dorsalis pedis artery; the two systems converge at the central third but do not form a rich anastomotic network, so the central third has the lowest vascular density in the bone. The medial third, which articulates with the medial cuneiform and takes the tibialis anterior insertion, and the lateral third, which articulates with the lateral cuneiform and contributes to cuboid stability, both have robust healing potential. The central third bears the compressive load at the apex of the medial longitudinal arch, and it is where the stress fracture, and the delayed union or nonunion, occur.

What the anatomy actually shows. The dorsal-central watershed geography is real, but McKeon's cadaveric injection study found a frankly avascular central third in only 11.8% of specimens; most had a dense intraosseous supply throughout, and the authors concluded that biomechanical factors, the high compressive load at the watershed during sprinting and jumping, may matter more than avascularity. Manage the central-third fracture as high-risk because of the load and the fracture pattern, not on an assumption of universal avascularity.

The load. The navicular is the keystone of the medial longitudinal arch and the central third is the apex of bending: the compressive force transmitted through the talar head meets the counterforce from the cuneiforms during running and jumping. Ground reaction force is 2-3 times body weight in running and higher in jumping, and it is the cumulative impact cycles of running, jumping and cutting that do the damage. A cavovarus foot overloads the lateral column and concentrates stress at the navicular.

Why repair loses. Repetitive impaction of this hypovascular zone outstrips repair. The dorsal cortex is thinner than the plantar in many specimens, intramedullary pressure is relatively low, which limits endosteal healing, and cortical bone mounts only a limited periosteal callus. Microdamage accumulates, and because the biology of repair is limited even a recognised fracture heals slowly, while unrecognised fractures progress to complete or displaced patterns.

The fracture pattern. The fracture begins on the dorsal cortex, the compression side of the bending moment, and runs in the sagittal plane, propagating from proximal to distal across the central third; the sagittal direction reflects the orientation of the watershed zone. A short, dorsal, partial fracture is the hallmark of an early stress injury.

Saxena's classification is based on CT appearance and is the most useful treatment-stratification tool. Both feet are usually scanned, and the contralateral side is often used to plan screw length. MRI grading and clinical staging complement it: the athlete who can no longer train, runs with a limp, or has pain at rest has crossed from a stress reaction into an established or complete fracture.

View larger

Vague, activity-related midfoot pain in a running, jumping or cutting athlete must prompt an N-spot examination, a hop test and MRI even when the X-ray looks normal.

History. The onset is insidious and gradual, often weeks to months before presentation. The pain is a vague midfoot or dorsal midfoot ache, worse with push-off, that settles with rest and recurs when training resumes. Ask about the sport and the training load (sprinting, jumping, cutting, a sudden increase in mileage) and about prior treatment: many have already had physiotherapy, orthotics or imaging.

The N-spot. The dorsal aspect of the proximal navicular, palpated with the thumb between the tibialis anterior and extensor hallucis longus tendons. Tenderness here is positive in almost all cases, and firm thumb pressure reproduces the patient's pain in the majority. Tenderness is occasionally bilateral, so compare with the contralateral side.

Provocation. The single-leg hop test, hopping in place on the affected foot, reproduces the pain in nearly every symptomatic athlete and is the most useful provocative bedside test. Standing on tiptoe also reproduces the pain. In established cases the gait is antalgic or the push-off shortened, and foot posture should be assessed for cavovarus, planovalgus and limb-length discrepancy.

The standard sequence in a high-performance athlete with vague midfoot pain is X-ray, then MRI if the X-ray is normal, then CT if the MRI is positive. Skipping the CT leaves the fracture ungraded and the need for fixation uncertain; skipping the MRI in a high-risk scenario (a high-performance sprinter, persistent pain despite non-weight-bearing) risks missing the diagnosis entirely.

Radiographs. AP, lateral and oblique views of the foot. Look for a sclerotic line in the central third, a cortical break on the dorsal aspect, cyst formation and, in chronic cases, a dorsal exostosis. Films are cheap and available and detect the established fracture, but they miss over 80 percent of early fractures: they are often normal for the first 4-6 weeks and normal in 70-80 percent of acute navicular stress fractures, so never rely on a normal X-ray to exclude the diagnosis.

MRI. The first imaging study of choice in a symptomatic athlete with normal X-rays. Fat-suppressed T2 or STIR sequences detect marrow oedema and T1 detects the fracture line; the findings are marrow oedema in the central third and a sagittal-plane fracture line on T1. Sensitivity approaches 100 percent for stress fracture and MRI detects the bone stress reaction, but it cannot always depict the full extent of the fracture, so a positive MRI is always followed by CT to classify by Saxena and plan management.

CT. Fine-cut (1 mm) CT in the plane of the navicular body with coronal and sagittal reformats. It shows the dorsal-cortex break, the extent of the fracture, displacement, sclerosis at the margins (nonunion) and cyst formation, and it is the best study for fracture morphology and for union. It cannot show oedema and it carries a radiation dose; its job is to grade the fracture and drive the decision between non-operative and operative management.

Other modalities. Ultrasound shows peritendinous oedema and the tibialis posterior tendon but has limited value for the navicular itself, an adjunct rather than a primary test. Bone scan is highly sensitive but poorly specific with no anatomic detail, and has largely been replaced by MRI.

Bone health. Recurrent stress fractures, features of the female athlete triad or a low BMI call for a DEXA scan, vitamin D, calcium, parathyroid hormone, oestradiol and a menstrual history, to identify and address the systemic component of the stress injury.

Activity-related midfoot or medial arch pain in an athlete has a broad differential, and the navicular stress fracture is the diagnosis not to miss. The named alternatives each have their own page where relevant: Accessory Navicular, Lisfranc injuries, Tarsal Coalition and Kohler Disease.

The algorithm. Treatment is stratified by Saxena type on CT. A Type I incomplete dorsal-cortex fracture is treated with a non-weight-bearing cast for 6-8 weeks; Type II and III fractures go to headless compression screw fixation from lateral to medial, often with bone graft; and return to sport is 4-6 months either way. That reflects widespread practice, but it is not a demonstrated superiority, and the strongest synthesis available points the other way.

What the evidence actually shows. Torg's meta-analysis (Am J Sports Med 2010;38(5):1048-53, PMID 20197494) pooled the published series into three strategies and found no statistically significant difference between non-weight-bearing conservative treatment and surgery (P = 0.6441), with a trend favouring NWB: 96% successful outcomes against 82% for surgery. Its stated conclusion is that non-weight-bearing management is the standard of care for both partial and complete stress fractures. What the same analysis establishes beyond doubt is that weight-bearing "rest" is inferior to both (P = 0.0001 versus NWB; P less than 0.0003 versus surgery), and that is the real error to avoid.

Saxena's own prospective series reported a return to activity of 4.0 months overall, with Type III at 4.2 months, and concluded that these fractures can take four months to heal with nonoperative or operative treatment, a null result on treatment. The earlier retrospective figure of 3.1 months for surgery against 4.3 for conservative care came from nine versus thirteen patients in whom fracture type and treatment were inseparable, because Type I fractures were significantly more likely to be managed conservatively.

The defensible position. Strict non-weight-bearing immobilisation is the default for every type, including complete fractures. Fixation is a reasonable option, not an obligation, for the Type III or sclerotic fracture, the established non-union, and the athlete who has already failed conservative treatment or for whom the certainty of a construct is worth the operative risk. Present it as a choice with no proven outcome advantage, not as the required treatment.

Return to sport. Once the cast is off, cast-treated and fixed fractures follow the same graded progression, and the same gate applies to both:

• Weeks 8-12: protected weight-bearing, low-impact cross-training (swimming, cycling); after fixation, intrinsic foot, calf and tibialis posterior strengthening begin here
• Weeks 12-16: walk-jog progression on flat ground; after cast treatment, intrinsic foot strengthening begins here
• Weeks 16-24: graduated return to running, sport-specific drills
• Full sport at 4-6 months: pain-free hop, no N-spot tenderness, CT-confirmed union

Approach. A limited open or percutaneous lateral incision over the lateral navicular. For a complete fracture without nonunion, percutaneous fixation is reasonable; an established nonunion or sclerosis usually requires open debridement and grafting. The open cases, a comminuted fracture, a nonunion requiring debridement or a failed percutaneous fixation, are exposed through a medial or dorsomedial approach.

Preparation and reduction. Curette the fibrous tissue at the fracture site and debride sclerotic margins back to bleeding bone, then reduce anatomically under direct vision and fluoroscopy before any drilling.

The screw. Pass the guidewire from lateral to medial across the central third along its long axis, perpendicular to the fracture and central in both planes, and confirm its position on AP, lateral and oblique views. Medial-to-lateral is avoided because of the risk to tibialis anterior and the saphenous nerve. Over the wire goes a cannulated headless compression screw of 3.0 to 4.5 mm (commonly 4.0-4.5 mm) with the threads crossing the fracture, compressing it by lag technique or with a partially threaded screw; single or double, the double screw increasingly used for rotational control; in the open case a mini-fragment plate is the alternative. Plan the length from the contralateral foot, foot length being reasonably symmetric, but confirm it on CT or fluoroscopy rather than assume.

Graft. Cancellous autograft from the distal tibia or calcaneus for an established nonunion, for sclerotic or cystic margins, or to fill a defect; do not rely on the screw alone in an established nonunion.

Before closing. Confirm the reduction and the hardware position fluoroscopically, and check that no screw penetrates a joint.