Evidence Summary

 Epidemiology of humeral shaft fractures
 Treatment of humeral shaft fractures
 Prognosis of humeral shaft fractures
 Recommendations for future research
 Further humerus shaft evidence material
Epidemiology of humeral shaft fractures
  • Fractures of the humeral shaft (diaphysis) account for approximately 3% of all fractures and represent 20% of all humeral fractures.
  • Primary causes of humeral fracture include traffic accident, accidental falls, or violent injury.
  • Thirty-three to 90% of humeral shaft fractures are reported  to be treated conservatively.  
  • Operative intervention may include compression plate fixation, intramedullary nailing, or external fixation, depending on the properties of the fracture and other associated injuries.
  • Perioperative complications of concern include iatrogenic comminution or radial nerve damage, nonunion and infection.1



Treatment of humeral shaft fractures

1. Functional bracing versus U-cast

  • No statistically significant differences between the use of functional bracing and U-casts were found when comparing nonunions or delayed unions.
  • Elbow extension was significantly greater after bracing; however, varus deformity of 5° was significantly more common with the brace when compared with U-cast treatment.


2. Functional bracing versus intramedullary nails

  • Free mobility of the shoulder was significantly greater in the functional brace group.
  • Limited external rotation was significantly greater in the nail group.
  • No significant differences were found regarding limited elevation, limited abduction, axial deviation, lateral displacement, or pain.


3.Functional Bracing versus plate fixation

  • Normal range of motion was achieved in 100% of patients in both groups
  • Plate fixation was associated with more instances of iatrogenic radial nerve palsy.


4. Intramedullary nails versus plates

  • Rigid intramedullary nails versus plates
    • Re-operation risk for IMN was almost three times that for plates based on pooled estimates for two studies. For every seven patients treated with plates, one re-operation could be avoided, NNT =7 (4-70).
    • Similar percentages of patients had nonunion, infection, and iatrogenic nerve injury in the two treatment groups, based on pooled estimates.
    • Shoulder pain and decreased shoulder range of motion was significantly less in patients treated with plates according to one study; another found no differences in pain or disability between treatments.
    • Time to healing was greater in plate groups than nail groups was not statistically significant.
  • Flexible or semi-rigid nail versus plates
    • Nonunions, infection and iatrogenic nerve palsy did not differ significantly between treatment methods in two studies


5. Antegrade versus retrograde insertion

  • Nonunion after nailing was similar between both insertion methods.
  • Retrograde insertion resulted in smaller differences in Constant’s Shoulder score when compared to uninjured control side.
  • No significant differences between treatment in measures of pain or complications were observed.


6. Ender nail versus Interlocking nail

  • Operation time and blood loss were significantly greater among interlocking nail surgeries.
  • Comparisons of nonunions, iatrogenic nerve palsy, intraoperative communition, and requirements for second operations did not find significant differences between nails.


7. ORIF with bone graft versus ORIF without bone graft

  • Nonunions were more common in the patients treated without bone graft.
  • Time to union was significantly longer in the non-bone graft group.
  • Average blood loss was significantly greater in the bone graft group, while operation time was significantly shorter when compared to the non-bone graft group.



Prognosis of humeral shaft fractures
  • Nonunion was associated with long oblique fracture pattern, alcohol abuse8, 10, and obesity.8, 10
  • Comminuted or open fractures, transverse fractures, those treated early with open reduction, and fractures occurring in the middle third of the humeral shaft are also more prone to nonunion.10
  • Alcoholism and chronic lung disease likely contribute to poor healing outcomes.10



Recommendations for future research
  • Randomized controlled trials or prospective cohort studies designed to reduce bias, misclassification and measurement error to compare operative treatments and surgical approaches based on fracture patterns.
  • Prospective prognostic studies, designed to reduce bias and measurement error and adjust for potentially confounding factors, should be conducted. 
  • Increased use of rigorously validated measures and attention to complete and standardized follow-up of all patients. 
  • More randomized controlled trials could add to the knowledge base for treatment of these fractures, perhaps involving multiple centers in order to accrue sufficient numbers of participants per treatment arm.
  • Additional comparisons between types of nails and insertion methods would be extremely useful, although since choice of hardware and method of surgery are often determined by type of fracture, comparisons may be challenging.