•Acute kidney injury (AKI) is common (61%) in critically-ill obstetric patients.•AKI increased duration of hospitalization and mortality (33% versus 0%).•A combination of urinary biomarkers TIMP-2 · IGFBP7 did not predict AKI.•A combination of urinary biomarkers TIMP-2 · IGFBP7 did not predict mortality.•Previous results for urinary TIMP-2 · IGFBP7 need validation in obstetrics. There are few data regarding acute kidney injury in critically-ill obstetric patients. A combination of urinary cell cycle arrest markers, tissue inhibitor of metalloproteinase-2 (TIMP-2) and insulin-like growth factor binding protein7 (IGFBP7), is validated for the early prediction of acute kidney injury in non-obstetric patients. We evaluated the epidemiology of acute kidney injury in critically-ill obstetric patients and the role of the biomarker combination in predicting acute kidney injury and mortality. Acute kidney injury, its severity and risk factors, were assessed using Kidney Disease: Improving Global Outcomes (KDIGO) guidelines during the intensive care unit stay. An ELISA technique measured TIMP-2 and IGFBP7 in urine samples collected at the time of admission there. Results for 66 patients showed an overall incidence of acute kidney injury of 40/66 (61%), with 50%, 10% and 40% being in stage 1, 2 and 3 respectively. Patients with acute kidney injury showed significantly greater sepsis and shock; longer stay and higher mortality during intensive care (33% vs 0%) and in hospital (38% vs 0%) compared to those without (P <0.05). The area-under-the receiver operating characteristics curve was <0.5 for urinary TIMP-2·IGFBP7 as a predictor of kidney injury and mortality (P >0.05). Acute kidney injury is common in critically-ill obstetric patients, increasing mortality and duration of hospitalization. It was significantly more common in patients with septic shock. Previously validated results of urinary TIMP-2·IGFBP7 that successfully predict early acute kidney injury or mortality are not applicable to obstetric patients.