Diabetes: foot ulcers: osteomyelitis: raised ESR and exposed bone were helpful.

Clinical bottom line (level 1b)

  1. Clinical suspicion, bone exposed within the ulcer, or an ESR > 100 made osteomyelitis far more likely.
  2. Ulcer size > 2sq cm or bone palpable in the ulcer made osteomyelitis more likely.
  3. X-ray studies were not clearly helpful.
  4. Osteomyelitis was difficult to rule out - probing the ulcer for bone may be useful (LR-0.63) .
Newman et al: Journal of the American Medical Association 1991; 266: 1246-1251
Grayson et al: Journal of the American Medical Association 1995; 273: 721-723
Expires August 2004

The study

Setting: tertiary medical centres

110 patients (aged study 1: mean 55 years; study 2: mean 60 years, ?% male) study 1: 35 diabetic patients with diabetes mellitus and 41severe foot ulcers, mean ulcer duration 4 months; study 2: 75 patients with diabetes mellitus and 76 foot ulcers and severe, limb-threatening foot infection

Excluded if
  • severe peripheral vascular disease
  • antibiotics for more than seven days
  • incomplete bone biopsy
  • nonhealed surgical wounds or exposure of the adjacent bone during debridement



    • Independent blinded reference standard, applied in all patients from a consecutive appropriate spectrum.
    Reference standard:
      • positive bone culture
      • positive histological examination on bone biopsy: by surgical debridement, amputation, or 15-gauge trocar placed through 5 mm incision in an area non-continuous with the foot ulcer in study. In study 2, bone biopsies were only done if the bone was palpable, exposed during debridement, or resected. Otherwise x-rays or appearance at surgery was used
    Diagnostic test: clinical findings (bone was probed with a sterile blunt steel eye-probe for palpable bone (hard-rock structure at the base of the ulcer)

    The evidence

    pre-test probability of osteomyelitis: 68%, (95% CI: 60% to 77%)

    diagnostic test osteomyelitis no osteomyelitis LR+
    (95% CI)    		 post-test probability LR-
    (95% CI)    		 post-test probability
    clinical suspicion 13 0 inf
    (2.1 to inf)     		100% 0.83
    (0.75 to 0.92)     		64%
    ulcer area >2 cm² 23 3 3.6
    (1.2 to 11)     		88% 0.76
    (0.63 to 0.91)     		62%
    bone exposed within ulcer 13 0 inf
    (2.1 to inf)     		100% 0.83
    (0.75 to 0.92)     		64%
    bone palpable in ulcer (2) 33 4 3.9
    (1.5 to 10)     		89% 0.63
    (0.50 to 0.80)     		58%
    x-ray 11 3 1.7
    (0.51 to 5.7)     		79% 0.93
    (0.81 to 1.1)     		67%
    ESR >100 8 0 inf
    (1.3 to inf)     		100% 0.89
    (0.83 to 0.97)     		66%
    bone scan 28 25 0.52
    (0.36 to 0.75)     		53% 2.2
    (1.3 to 3.8)     		82%
    24 hr leukocyte scans 31 11 1.3
    (0.75 to 2.3)     		74% 0.86
    (0.64 to 1.2)     		65%
    total 75 35

    Comments

    1. Results were combined from two studies.

    Citation

    1. Newman LG, Waller J, Palestro CJ, et al: Unsuspected osteomyelitis in diabetic foot ulcers. Diagnosis and monitoring by leukocyte scanning with indium In 111 oxyquinoline. Journal of the American Medical Association 1991; 266: 1246-1251
    2. Grayson , et al: Probing to bone in infected pedal ulcers. A clinical sign of underlying osteomyelitis in diabetic patients. Journal of the American Medical Association 1995; 273: 721-723
    Contributor: Chris Ball and Clare Wotton, August 2000
    Reviewer:

    Clinical Question.
    Patient diabetic foot ulcer
    Intervention or Exposure clinical findings
    Outcome osteomyelitis