- Design
- systematic review with random-effects meta-analysis; 13 studies reviewed, five in the primary pooled analysis
- Population
- 796 children and adolescents with type 1 diabetes using insulin pumps and related diabetes technology
- Primary outcome
- pooled prevalence of any device-related skin complication, and of specific complication types
- Effect
- any complication 63.8% (95% CI 47.8-77.2, I-squared 92.1%); scarring 36.7%, lipohypertrophy 24.1%, eczema-like 16.6%, infection 8.6%, lipoatrophy 5.1%
As continuous insulin delivery becomes standard in paediatric type 1 diabetes, the skin under the devices has become a dermatological problem that diabetes clinics are not set up to look for. This review pooled five studies covering 796 children and adolescents for its primary estimate.
Any skin complication had a pooled prevalence of 63.8% (95% CI 47.8 to 77.2). Broken down: scarring or insertion marks in 36.7%, lipohypertrophy in 24.1%, eczema or dermatitis-like reactions in 16.6%, infection or folliculitis in 8.6%, and lipoatrophy in 5.1%. Heterogeneity was very high at 92.1%.
That heterogeneity, and the authors' own caution about inconsistent definitions and ascertainment, mean the 63.8% figure should be read as 'very common' rather than as a rate. But the clinical implication does not depend on the precise number. Lipohypertrophy at a quarter of children matters because it changes insulin absorption and therefore glycaemic control — a child with unexplained variable glucose readings may have a palpable problem at the infusion site rather than a behavioural one. And the eczematous reactions are treatable dermatological problems that are currently being managed, if at all, by diabetes teams. Asking to examine the sites, at the diabetes review and in the dermatology clinic, is the whole intervention.
- Examine and palpate infusion and sensor sites at every diabetes review, not only when the family complains
- Suspect lipohypertrophy in a child with unexplained glycaemic variability before assuming non-adherence
- Treat adhesive-related dermatitis as a dermatological problem — barrier films, adhesive changes and topical treatment all help
- Reinforce site rotation explicitly; scarring and insertion marks affected over a third
- The pooled prevalence is imprecise (I-squared 92.1%); use it to justify looking, not to quote a rate to families
The statistics, in plain English
An I-squared of 92.1% means the five studies disagreed almost entirely about how common these complications are, so the pooled 63.8% is an average across studies that were measuring different things in different ways — the confidence interval of 47.8 to 77.2 reflects that disagreement rather than sampling error alone. Prevalence estimates are also highly sensitive to whether investigators actively examined the skin or relied on reported problems; active examination finds far more. The individual complication percentages come from varying subsets of studies and are correspondingly less certain.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for dermatology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free