REF-1379 · Hoenig et al., 2023 · British Journal of Sports Medicine · Paper · PubMed

How long until I can run again after a stress fracture?

It depends on the bone: pooled study means ran from 44 days (posteromedial tibia) to 127 (navicular), with wide ranges, and over 90% returned. Averages, not your date.

Main caveat: Read from the abstract only. The pooling model, heterogeneity, risk of bias, the definition of return to sport and of a complication, and the times for all…
fairly strong evidenceMeasured on a mixed group

Hoenig and colleagues asked whether the usual clinical split of bone stress injuries into low- and high-risk sites, which the abstract says "has not been sufficiently validated", holds up against outcomes.

They searched PubMed, Web of Science, Cochrane CENTRAL and Google Scholar from inception to December 2021 for peer-reviewed studies reporting site-specific return to sport after bone stress injuries of the lower extremity and pelvis in athletes, and included 76 studies with 2,974 injuries.

Of the 16 studies that compared several sites, 11 described the anatomical site as prognostic for return to sport or for the rate of treatment complication.

Pooled, return took longest after injuries of the tarsal navicular (127 days; 95% CI 102 to 151) and femoral neck (107 days; 79 to 135) and was shortest after the posteromedial tibial shaft (44 days; 27 to 61) and fibula (56 days; 13 to 100); overall more than 90% of athletes returned to sport.

Treatment complications were most frequent at the femoral neck, tarsal navicular, anterior tibial shaft and fifth metatarsal and least frequent at the fibula, pubic bone and posteromedial tibial shaft.

The authors conclude that site influences both the timeline and the risk of complication, with the femoral neck, anterior tibial shaft and tarsal navicular the more difficult cases.

Our reading: this is the first entry in the base that puts numbers on "it depends on the bone" — about six weeks for the posteromedial tibial shaft against about four months for the navicular, on average across studies; the intervals are wide (the fibula's runs from about two weeks to over three months), and the abstract does not give the times for the other sites, the definition of a return or the complication rates themselves.

Protocol

Design: systematic review and meta-analysis, registered (PROSPERO CRD42021232351).
Sources: PubMed, Web of Science, Cochrane CENTRAL and Google Scholar, from database inception to December 2021.
Eligibility: peer-reviewed studies that reported site-specific return to sport after a bone stress injury of the lower extremity or pelvis in athletes.
Included: n = 76 studies reporting on n = 2,974 bone stress injuries; 16 studies compared several injury sites.
Outcomes: time to return to sport by site; rate of treatment complication by site. How return to sport and complication were defined, the pooling model, heterogeneity and risk of bias are not in the abstract.
Funding and interests: the abstract states no funding; Crossref funder metadata lists the National Institutes of Health (award NIH/NIAMS P30 AR072581). Competing interests as printed in the PubMed record: "KEA is a deputy editor of BJSM." No device maker, federation or data platform is named.

Key Results

SitePooled time to return to sport (95% CI)Complication rate (abstract's ranking)
Tarsal navicular127 days (102 to 151)among the highest
Femoral neck107 days (79 to 135)among the highest
Fibula56 days (13 to 100)among the lowest
Posteromedial tibial shaft44 days (27 to 61)among the lowest
Anterior tibial shaftnot in the abstractamong the highest
Fifth metatarsalnot in the abstractamong the highest
Pubic bonenot in the abstractamong the lowest
Other sites (e.g. sacrum, calcaneus, other metatarsals)not in the abstractnot in the abstract
Returned to sport overall: more than 90% of athletes.
Site as a prognostic factor: 11 of the 16 studies that compared several sites described the site as prognostic for return to sport or for complications.
Not in the abstract: complication rates as numbers, what counted as a complication, return times for any site beyond the four above, and any split by sex, sport or imaging grade.

Mechanisms

Measured (pooled): return-to-sport time and complication frequency differed by the bone injured, with the navicular and femoral neck slowest and the posteromedial tibia and fibula fastest.
Authors' interpretation: the anatomical site of a bone stress injury influences return-to-sport timelines and the risk of complication; femoral neck, anterior tibial shaft and tarsal navicular injuries are the more challenging ones. The abstract frames this as support for the low- and high-risk classification, which until now rested on clinical practice.
Inference: a bone stress injury is repeated loading outpacing repair (PHYS-37); why some sites heal more slowly or fail more often is not explained in the abstract, so the entry gives no mechanism for the site differences.
Inference: the four times are means of study means. A mean of 44 days for the posteromedial tibia says that roughly six weeks is typical across studies, not that six weeks is safe for a given runner; the wide intervals (the fibula's spans 13 to 100 days) reflect few or small studies at some sites as much as real spread between athletes.

Practical Application

How long until I can run again after a stress fracture? It depends on the bone: pooled study means ran from 44 days (posteromedial tibia) to 127 (navicular), with wide ranges, and over 90% returned. Averages, not your date.
Do not promise a date; give the range for the site the clinician named: about six weeks on average for the posteromedial (inner-back) tibial shaft (44 days, 95% CI 27 to 61) and about two months for the fibula (56 days, 13 to 100), against about three and a half to four months for the femoral neck (107 days, 79 to 135) and the tarsal navicular (127 days, 102 to 151) (A-, from the abstract).These are averages across studies of athletes; the treating clinician's criteria, not the calendar, decide the restart (REF-3498 for the tibial criteria; REF-1383 for the low-risk tibial and metatarsal load rule).
Some sites deserve more caution from the start: the femoral neck, tarsal navicular, anterior tibial shaft and fifth metatarsal had the most treatment complications, the fibula, pubic bone and posteromedial tibial shaft the fewest. Not from this paper; the base's own practice: bone-like pain at any site means no running and a clinician (SYNTH-73 red flags); the site ranking is a reason to say so more firmly, not a licence to run on a "low-risk" bone without a diagnosis.
Red-flag sites, sharpened (inference about where each site's pain is felt): SYNTH-73's list names groin, front-of-shin or forefoot pain; this review's complication-prone sites map onto groin or front-of-hip pain (femoral neck), front-of-shin pain (anterior tibial shaft) and outer-forefoot pain (fifth metatarsal), and add the top of the midfoot (navicular), which the list does not name.
What it is good for as a source: the size of the difference between sites, and the ranking of sites by complications, across 76 studies. What it does not show: a return date for one person, return times for the sites not reported in the abstract, how return was defined or counted, what the complications were or how often they happened, whether women and men or runners and other athletes differ, or which treatment shortens the return.

Limitations

Read from the abstract only. The pooling model, heterogeneity, risk of bias, the definition of return to sport and of a complication, and the times for all sites but four could not be checked.
Designs not given, almost certainly observational. The abstract names no trial and does not give the study designs (inference: case series and cohorts); only 16 of 76 studies compared sites within one sample, so most site estimates come from different studies, and site differences may partly reflect how different centres treated and counted, not only the bone.
Wide intervals. The fibula's 95% CI runs from 13 to 100 days; any single figure is a midpoint of a broad range.
Who. "Athletes" of unstated sports, levels, ages and sex; runners are the base's use case but not the stated population.
Interests and funding: competing interests as printed in PubMed: one author (KEA) is a deputy editor of the journal. The abstract states no funding; Crossref lists the National Institutes of Health (NIAMS P30 AR072581). No device maker, federation or data platform is named.
How this entry was checked and changed
Why this level

(from the abstract) — a systematic review with pooled estimates, held below A because the abstract describes no included study as a trial (the designs are not given; a time-to-return outcome by site is typically reported by case series and cohorts — inference), only 16 of the 76 studies compared more than one site, and the abstract gives no risk-of-bias rating, heterogeneity statistic, definition of return to sport or definition of a treatment complication. Read from the abstract (closed access; Europe PMC holds no full text and OpenAlex lists no open copy)