Cellulitis: lymphoedema and a site of entry increased the risk of cellulitis.

Clinical bottom line (level 3b)

  1. Patients were at increased risk of cellulitis if they:
    • had lymphoedema (NNH = 15 at unknown)
    • had a site of entry (NNH = 45 at unknown)
    • had leg oedema (NNH = 670 at unknown)
    • had venous insufficiency (NNH = 530 at unknown)
    • were overweight (NNH = 1000 at unknown)
Dupuy et al: British Journal of Medicine 1999; 318: 1591-1594
Expires November 2004

The study

Case-control study with unblinded, unobjective outcomes, adjusted for confounding factors, not validated in an independent set of patients.

Setting: seven hospitals, France

461 patients (aged ?, ?% male) cellulitis

Excluded if
  • <15 years old
  • abscess or necrotising fasciitis


    • Cases: 167 patients (52% male, mean age 57): cellulitis of the leg, defined as sudden onset (<24 hours) of a well demarcated cutaneous inflammation, with fever >38 ° C or chills.
    Controls: 294 patients (52% male, mean age 57): matched for age, sex and hospital- admitted for an acute condition not a priori related to one of the suspected risk factors nor related to a chronic disease. Admitted for trauma, dermatological conditions, abdominal surgery, infection, orthopaedic surgery, vascular disease, sciatalgia, eye disease and other conditions

    Factors studied:
  • cellulitis


    • Factors summarised:
  • lymphoedema
  • site of entry presence of leg ulcer, wound, excoriated leg dermatosis, pressure ulcer, toe-web intertrigo
  • leg oedema
  • venous insufficiency presence of at least one of: history of either venous leg ulcer or leg phlebitis, or specific dermatitis
  • overweight (>120% of Lorentz's formula)


    • Multivariate analysis was used to adjust for confounding factors.

    Outcomes studied:
  • cellulitis

    • The evidence

    Patient expected event rate for cellulitis: 0.1%
    risk factor for
    cellulitis     		adjusted OR
    (95% CI)    		 NNH
    (95% CI)
    lymphoedema 71.2
    (5.6 to 908)     		15
    (2 to 220)
    site of entry 23.8
    (10.7 to 52.5)     		45
    (20 to 100)
    leg oedema 2.50
    (1.2 to 5.1)     		670
    (250 to 5000)
    venous insufficiency 2.90
    (1.0 to 8.7)     		530
    (130 to infinity)
    overweight 2.00
    (1.1 to 3.7)     		1000
    (370 to 10000)

    Comments

    1. The use of a single interviewer describing the risk factors present in patients who have an obvious disease (cellulitis) increases the risk of bias. A better system would have used two observers and calculated an agreement.

    Citation

    1. Dupuy A, Benchikhi H, Roujeau J-C, et al: Risk factors for erysipelas of the leg (cellulitis): case-control study. British Journal of Medicine 1999; 318: 1591-1594
    Contributor: Clare Wotton and Bob Phillips, November 2000
    Reviewer: Santiago Alvarez Montero

    Clinical Question.
    Patient patients
    Intervention or Exposure risk factors
    Outcome cellulitis