Research Article: Aggressive fluid accumulation is associated with acute kidney injury and mortality in a cohort of patients with severe pneumonia caused by influenza A H1N1 virus

Date Published: February 15, 2018

Publisher: Public Library of Science

Author(s): Gustavo Alejandro Casas-Aparicio, Isabel León-Rodríguez, Rafael de Jesús Hernández-Zenteno, Manuel Castillejos-López, Claudia Alvarado-de la Barrera, Christopher E. Ormsby, Gustavo Reyes-Terán, Emmanuel A Burdmann.


Fluid accumulation is associated with adverse outcomes such as acute kidney injury (AKI) in critically ill patients. This study aimed to describe the factors associated with AKI in individuals with influenza A H1N1 severe pneumonia, and explore the relation of fluid accumulation with AKI and mortality.

We reviewed medical records of individuals with influenza A H1N1 severe pneumonia and no history of chronic kidney disease, attending a national referral center for respiratory diseases between November 2014 and May 2015. Demographic information, risk factors for AKI, physiologic and laboratory data, outcomes and information on fluid intake and output were recorded. Categorical variables were compared using the chi-square test. Quantitative variables were compared using the Mann-Whitney test. Factors associated with AKI and mortality were identified by binary logistic regression. Linear models of fluid accumulation rates for individuals and groups were estimated using segmented linear regression.

Of 60 patients studied, 43 developed AKI (71.6%). Male gender was protective for AKI (p = 0.019). AKI was associated with nephrotoxic drugs (p = 0.016); PEEP>10 cm H2O on admission (p = 0.031); mortality (p = 0.037); and fluid accumulation ≥10% (fluid overload) at day 7 of hospitalization (p = 0.00026). Mortality was associated with older age (p = 0.009); nephrotoxic drugs (p = 0.034); and higher Pneumonia Severity Index score (112 vs. 76, p = 0.008) on admission. The Deceased-AKI group had a higher rate of fluid accumulation (expressed as ml/kg/body weight) than the Survivors-No AKI group during the study period of 7 days (Survivors-No AKI = 13.31 vs. Deceased-AKI = 22.76, p = 0.019). During the highest phase of fluid accumulation, the Survivors-No AKI group had a slower rate of fluid accumulation than the Survivors-AKI group (14.91 vs. 28.49, p = 0.001).

A high rate of fluid accumulation was associated with AKI and mortality. We support the approach of resuscitation in acute illness, with an early transition to neutral and then negative fluid balances.

Partial Text

Influenza A H1N1 virus infection is associated with a spectrum of illnesses, ranging from upper respiratory infection, to multiple organ dysfunction and death. In critical patients with severe influenza A H1N1 disease, acute kidney injury (AKI) is a common complication. AKI is associated with development of chronic kidney disease [1], increased mortality, adverse outcomes and longer periods of intensive care unit stay [2, 3]. In most cases, AKI develops from a combination of factors including hypovolemia, sepsis, nephrotoxins and hemodynamic perturbations [4]. In order to restore cardiac output, systemic blood pressure and renal perfusion in critically ill patients, an adequate fluid resuscitation is necessary [5]. However, fluid administration beyond the correction of hypovolemia is associated with AKI, longer periods of hospital stay, increased mortality [6], organ dysfunction and worse clinical outcomes [7]. Fluid overload (FO) may derive from the combination of oliguria and fluid administration, leading to a positive fluid balance [8, 9]. Due to the dichotomy between traditional teaching and evolving evidence, wide variations in clinical fluid management exist and traditional practice involving administration of large fluid volumes is being questioned [5]. Prolonged fluid resuscitation leads to edema in the kidneys and other organs. As an encapsulated organ, the kidney is particularly affected by fluid congestion and raised venous pressures with a disproportionate elevation of intracapsular pressure, which leads to a decrease in renal blood flow and glomerular filtration rate [10].

During the period between November 2014 and May 2015, 184 individuals were diagnosed with severe pneumonia at our institution. Of those, 124 were deemed ineligible for this study because they did not fulfill the inclusion criteria (8 were under 18 years of age; 22 had incomplete clinical files; 9 had non-H1N1 pneumonia; and on the remaining 85 individuals, either influenza A H1N1 could not be confirmed by rRT-PCR, or they had PaO2/FiO2 >200 mm Hg). We thus included 60 individuals in the study. Of those, 39 were men (65%). The median age was 47.5 years, IQR, 44.2–50.9; 56.7% were obese; 21.7% had systemic arterial hypertension; 8.3% had diabetes mellitus; 6.7% had pneumopathies; and 38.3% had a smoking history. Only 1.7% of this cohort had been vaccinated against influenza. Sudden onset of influenza symptoms (<72 hours) was observed in 63.3% of the cases. The median APACHE II score on admission was 13.5 (IQR, 9.25–17). Forty-three individuals developed AKI (the AKI group). Of those, 9 had AKI stage 1 (20.8%); 16 had AKI stage 2 (37.1%) and 18 had AKI stage 3 (41.7%). Four patients requiring replacement therapy (6.7%) received sustained low-efficiency dialysis, and all of them survived. We explored the risk factors associated with the development of AKI and the relation of the accumulated fluid balance on AKI and mortality, in a homogeneous population with primary acute respiratory distress syndrome caused by influenza A H1N1 virus infection. By using the KDIGO classification, we found that 71.6% of individuals developed AKI. This proportion is similar to that found in a Canadian cohort, of 66.7% using the RIFLE classification [3]. AKI had a significant association with the use of nephrotoxic drugs during hospitalization. In agreement with a previous study in critically ill patients with lung disease, we also found a significant association between AKI and PEEP level >10 cm H2O on admission [16]. Male gender had a protective effect for AKI. Other factors previously associated with AKI include vasopressor use, mechanical ventilation, high APACHE II score, severe acidosis, high levels of C-reactive protein and lactic dehydrogenase upon ICU admission, longer stays in the ICU [17], high body mass index, history of asthma [3], high SOFA score, and greater incidence of shock, multiorgan dysfunction syndrome and coinfection [18]. In our cohort, the AKI group had higher PSI, APACHE II and SOFA scores on admission, as well as longer hospital stays and mechanical ventilation periods than the non-AKI group, but these differences were non-significant. We found lower PaO2/FiO2 in the AKI than in the non-AKI group, as well as lower PaO2/FiO2 in deceased individuals than in survivors, but lack of statistical significance in this parameter was probably affected by the inclusion criteria of PaO2/FiO2 <200 mm Hg on admission. Our results support previous findings in critically ill patients, indicating that fluid overload and rapid fluid balance increase are associated with AKI and mortality. Scientific evidence obtained from randomized clinical trials exploring more rational and flexible approaches to fluid therapy is required for minimization of the adverse consequences of fluid overload.   Source:


0 0 vote
Article Rating
Notify of
Inline Feedbacks
View all comments