Background:

Relatively little is known about the impact of renal dysfunction on morbidity and mortality in patients with pneumonia, especially in non–severe disease. The contribution of intravenous computed tomography contrast dye to renal dysfunction in this group is also unclear.

Methods:

We reviewed the electronic medical record of 4,459 consecutive admissions of adults =18 years of age with the principal discharge diagnosis of pneumonia from a large, tertiary care hospital and nearby community hospital between March 1, 2007 and March 30, 2010. Patients were classified by RIFLE stage based on change in measured or estimated baseline creatinine in relation to peak creatinine. We additionally grouped patients into a non–RIFLE category of “pre–Risk” which we defined as an increase in creatinine of >0.3mg/dl but < 50% over baseline. Severe pneumonia was defined as requiring admission to a medical ICU during the hospital course.

Results:

Our findings are presented in the attached table. Odd ratios were adjusted for age, male sex, and baseline creatinine. Patients with AKI were older, included more males, and had a higher baseline creatinine compared to those without AKI. 1053 patients received IV contrast; these patients were more likely to be younger, female, and with lower baseline and initial creatinine values. Use of contrast was associated with higher rates of ICU admission, mechanical ventilation, higher ICU and hospital lengths of stay, and in–hospital mortality. Time to renal recovery was significantly longer in those receiving contrast. However, receipt of contrast did not appear to be associated with increased rates or severity of AKI.

Conclusions:

AKI was observed in nearly half of all patients and conferred higher morbidity and mortality, increasing progressively with RIFLE stage. Even small, incremental changes in serum creatinine were associated with greater morbidity and mortality. Subjects classified as pre–risk had longer hospital and ICU length of stays, higher rates of ICU admissions, and a greater need for mechanical ventilation, as compared to those without any rise in creatinine. Similar effects on morbidity and mortality were seen in the non–severe pneumonia cohort.

Figure 1Increasing Morbidity and Mortality with RIFLE Progression.