Study objective: To determine the sensitivity of the urine ketone dip test (UKDT) for the detection of ketonemia in patients with diabetic ketoacidosis (DKA) and diabetic ketosis (DK) in the ED.
Methods: We conducted a retrospective chart review in the ED of an urban, university-affiliated county teaching hospital. The study population comprised patients seen in the ED during 1994 and 1995 with a discharge diagnosis of DKA or DK and underwent urinalysis within 4 hours of the initial serum electrolyte and ketone determinations. We calculated test sensitivity, along with 95% confidence intervals (CIs).
Results: One hundred forty-eight patients with 223 occurrences diagnosed as DKA or DK were seen in the ED during the study period. One hundred fourteen patients with 146 occurrences of DKA or DK met all inclusion criteria; these patients made up the study group. There were 99 cases of DKA and 47 cases of DK. The sensitivity of the UKDT for the detection of ketonemia in all patients with DKA or DK was 97% (95% CI, 94% to 99%). In the subgroup of patients with DKA, the sensitivity of the UK was 97% (95% CI, 92% to 99%). For patients with DK, the sensitivity of the UK was 98% (95% CI, 89% to 99%).
Conclusion: The UKDT is highly sensitive for the presence of serum ketones in patients with DKA and DK. Prospective study is suggested to determine the specificity of the UKDT in this application and to validate its use as a screening tool for the detection of ketonemia in DKA and DK.