Acute kidney injury developed in the intensive care unit: a population-based prospective cohort study in the Brazilian Amazon

Abstract

The Brazilian Amazon is a vast area with limited health care resources. To assess the epidemiology of critically ill acute kidney injury (AKI) patients in this area, a prospective cohort study of 1029 adult patients of the three intensive care units (ICUs) of Rio Branco city, the capital of Acre state, were evaluated from February 2014 to February 2016. The incidence of AKI was 53.3%. Risk factors for AKI included higher age, nonsurgical patients, admission to the ICU from the ward, higher Acute Physiology and Chronic Health Evaluation (APACHE) II scores at ICU admission, and positive fluid balance > 1500 ml/24 hours in the days before AKI development in the ICU, with aOR of 1.3 (95% CI 1.03-1.23), 1.47 (95% CI 1.07-2.03), 1.96 (95% CI 1.40-2.74), 1.05 (95% CI 1.03-1.08) for each unit increase, and 1.62 (95% CI 1.16-2.26), respectively. AKI was associated with higher ICU mortality (aOR 2.03, 95% CI 1.29-3.18). AKI mortality was independently associated with higher age, nonsurgical patients, sepsis at ICU admission, presence of shock or use of vasoactive drugs, mechanical ventilation and mean positive fluid balance in the ICU > 1500 ml/24 hours, both during ICU follow-up, with aOR 1.27 (95% CI 1.14-1.43) for each 10-year increase, 1.64 (95% CI 1.07-2.52), 2.35 (95% CI 1.14-4.83), 1.88 (95% CI 1.03-3.44), 6.73 (95% CI 4.08-11.09), 2.31 (95% CI 1.52-3.53), respectively. Adjusted hazard ratios for AKI mortality 30 and 31-180 days after ICU discharge were 3.13 (95% CI 1.84-5.31) and 1.69 (95% CI 0.99-2.90), respectively. AKI incidence was strikingly high among critically ill patients in the Brazilian Amazon. The AKI etiology, risk factors and outcomes were similar to those described in high-income countries, but mortality rates were higher.

Keywords: Acute kidney injury; Amazon; Critically ill patients; Disadvantaged populations; Epidemiology; Low-income and middle-income countries.

Fig. 1

Survival curves for the first 30 days after discharge from the ICU. AKI^$ was evaluated in four different Cox regression models: (A) AKI development in the ICU; (B) AKI categorized by maximum KDIGO stage; (C) AKI categorized as transient (duration up to 48 h) or persistent (more than 48 h); and (D) AKI categorized as those developed up to or after the third day of ICU^# stay, using the absence of AKI as a reference. The Cox analyses were adjusted by age, nonsurgical patients, use of mechanical ventilation, use of vasoactive drugs or presence of shock, use of KRT, and use of diuretics during follow-up (first seven days of ICU stay), censored at 30 days. ^$AKI acute kidney disease, ^#ICU intensive care units.

Fig. 2

Survival curves for 31 to 180 days after discharge from the ICU for patients who survived the first 30 days. AKI^$ was evaluated in four different Cox regression models: (A) AKI development in the ICU; (B) AKI categorized by maximum KDIGO stage; (C) AKI categorized as transient (duration up to 48 h) or persistent (more than 48 h); and (D) AKI categorized as those developed up to or after the third day of ICU^# stay, using the absence of AKI as a reference. The Cox analyses of patients who survived the first 30 days after ICU discharge were adjusted by age, use of mechanical ventilation, use of vasoactive drugs or presence of shock, and use of KRT during follow-up (first seven days of ICU stay), censored at 180 days. ^$AKI acute kidney disease, ^#ICU intensive care units.