In observational studies, control of confounding can be done in the design and analysis phases. Using examples from large health care database studies, this article provides the clinicians with an overview of standard methods in the analysis phase, such as stratification, standardization, multivariable regression analysis and propensity score (PS) methods, together with the more advanced high-dimensional propensity score (HD-PS) method. We describe the progression from simple stratification confined to the inclusion of a few potential confounders to complex modeling procedures such as the HD-PS approach by which hundreds of potential confounders are extracted from large health care databases. Stratification and standardization assist in the understanding of the data at a detailed level, while accounting for potential confounders. Incorporating several potential confounders in the analysis typically implies the choice between multivariable analysis and PS methods. Although PS methods have gained remarkable popularity in recent years, there is an ongoing discussion on the advantages and disadvantages of PS methods as compared to those of multivariable analysis. Furthermore, the HD-PS method, despite its generous inclusion of potential confounders, is also associated with potential pitfalls. All methods are dependent on the assumption of no unknown, unmeasured and residual confounding and suffer from the difficulty of identifying true confounders. Even in large health care databases, insufficient or poor data may contribute to these challenges. The trend in data collection is to compile more fine-grained data on lifestyle and severity of diseases, based on self-reporting and modern technologies. This will surely improve our ability to incorporate relevant confounders or their proxies. However, despite a remarkable development of methods that account for confounding and new data opportunities, confounding will remain a serious issue. Considering the advantages and disadvantages of different methods, we emphasize the importance of the clinical input and of the interplay between clinicians and analysts to ensure a proper analysis.

IntroductionThe prognostic impact of acute kidney injury (AKI) on long-term clinical outcomes remains controversial. We examined the five-year risk of death, myocardial infarction, and stroke after elective cardiac surgery complicated by AKI.MethodsWe conducted a cohort study among adult elective cardiac surgical patients without severe chronic kidney disease and/or previous heart or renal transplant surgery using data from population-based registries. AKI was defined by the Acute Kidney Injury Network (AKIN) criteria as a 50% increase in serum creatinine from baseline level, acute creatinine rise of ≥26.5 μmol/L (0.3 mg/dL) within 48 hours, and/or initiation of renal replacement therapy within five days after surgery. We followed patients from the fifth post-operative day until myocardial infarction, stroke or death within five years. Five-year risk was computed by the cumulative incidence method and compared with hazards ratios (HR) from a Cox proportional hazards regression model adjusting for propensity score.ResultsA total of 287 (27.9%) of 1,030 patients developed AKI. Five-year risk of death was 26.5% (95% CI: 21.2 to 32.0) among patients with AKI and 12.1% (95% CI: 10.0 to 14.7) among patients without AKI. The corresponding adjusted HR of death was 1.6 (95% CI: 1.1 to 2.2). Five-year risk of myocardial infarction was 5.0% (95% CI: 2.9 to 8.1) among patients with AKI and 3.3% (95% CI: 2.1 to 4.8) among patients without AKI. Five-year risk of stroke was 5.0% (95% CI: 2.8 to 7.9) among patients with AKI and 4.2% (95% CI: 2.9 to 5.8) among patients without AKI. Adjusted HRs were 1.5 (95% CI: 0.7 to 3.2) of myocardial infarction and 0.9 (95% CI: 0.5 to 1.8) of stroke.ConclusionsAKI, within five days after elective cardiac surgery, was associated with increased five-year mortality and a statistically insignificant increased risk of myocardial infarction. No association was seen with the risk of stroke.

IntroductionAcute kidney injury (AKI) is common among intensive care unit (ICU) patients, but follow-up data on subsequent risk of cardiovascular disease remain sparse. We examined the impact of AKI on three-year risk of first-time heart failure, myocardial infarction (MI), and stroke among ICU patients surviving to hospital discharge, and whether this risk is modified by renal recovery before hospital discharge.MethodsWe used population-based medical registries to identify all adult patients admitted to an ICU in Northern Denmark between 2005 and 2010 who survived to hospital discharge and who had no previous or concurrent diagnosis of heart failure, MI, or stroke. AKI was defined according to the creatinine criteria in the Kidney Disease Improving Global Outcomes classification. We computed the three-year cumulative risk of hospitalization with heart failure, MI, and stroke for patients with and without AKI and the hazard ratios (HRs), using a Cox model adjusted for potential confounders.ResultsAmong 21,556 ICU patients surviving to hospital discharge, 4,792 (22.2%) had an AKI episode. Three-year cumulative risk of heart failure was 2.2% in patients without AKI, 5.0% for AKI stage 1, and 5.0% for stages 2 to 3. The corresponding adjusted HRs were 1.33 (95% confidence interval (CI), 1.06 to 1.66) for patients with AKI stage 1 and 1.45 (95% CI, 1.14 to 1.84) for AKI stages 2 to 3, compared to patients without AKI. The three-year cumulative MI risk was 1.0% for patients without AKI, 1.8% for patients with AKI stage 1 and 2.3% for patients with AKI stages 2 to 3. The adjusted HR for MI was 1.04 (95% CI, 0.71 to 1.51) for patients with AKI stage 1 and 1.51 (95% CI, 1.05 to 2.18) for patients with AKI stages 2 to 3, compared with patients without AKI. We found no association between AKI and stroke. The increased risk of heart failure and MI persisted in patients with renal recovery before discharge, although it was less pronounced than in patients without renal recovery.ConclusionsICU patients surviving any stage of AKI are at increased three-year risk of heart failure, but not stroke. Only AKI stages 2 to 3 are associated with increased MI risk.Electronic supplementary materialThe online version of this article (doi:10.1186/s13054-014-0492-2) contains supplementary material, which is available to authorized users.

IntroductionThere are few studies on long-term mortality among intensive care unit (ICU) patients with acute kidney injury (AKI). We assessed the prevalence of AKI at ICU admission, its impact on mortality during one year of follow-up, and whether the influence of AKI varied in subgroups of ICU patients.MethodsWe identified all adults admitted to any ICU in Northern Denmark (approximately 1.15 million inhabitants) from 2005 through 2010 using population-based medical registries. AKI was defined at ICU admission based on the risk, injury, failure, loss of kidney function, and end-stage kidney disease (RIFLE) classification, using plasma creatinine changes. We included four severity levels: AKI-risk, AKI-injury, AKI-failure, and without AKI. We estimated cumulative mortality by the Kaplan-Meier method and hazard ratios (HRs) using a Cox model adjusted for potential confounders. We computed estimates for all ICU patients and for subgroups with different comorbidity levels, chronic kidney disease status, surgical status, primary hospital diagnosis, and treatment with mechanical ventilation or with inotropes/vasopressors.ResultsWe identified 30,762 ICU patients, of which 4,793 (15.6%) had AKI at ICU admission. Thirty-day mortality was 35.5% for the AKI-risk group, 44.2% for the AKI-injury group, and 41.0% for the AKI-failure group, compared with 12.8% for patients without AKI. The corresponding adjusted HRs were 1.96 (95% confidence interval (CI) 1.80-2.13), 2.60 (95% CI 2.38 to 2.85) and 2.41 (95% CI 2.21 to 2.64), compared to patients without AKI. Among patients surviving 30 days (n = 25,539), 31- to 365 day mortality was 20.5% for the AKI-risk group, 23.8% for the AKI-injury group, and 23.2% for the AKI-failure group, compared with 10.7% for patients without AKI, corresponding to adjusted HRs of 1.33 (95% CI 1.17 to 1.51), 1.60 (95% CI 1.37 to1.87), and 1.64 (95% CI 1.42 to 1.90), respectively. The association between AKI and 30-day mortality was evident in subgroups of the ICU population, with associations persisting in most subgroups during the 31- to 365-day follow-up period, although to a lesser extent than for the 30-day period.ConclusionsAKI at ICU admission is an important prognostic factor for mortality throughout the subsequent year.

IntroductionDialysis-requiring acute kidney injury (D-AKI) is common among intensive care unit (ICU) patients. However, follow-up data on the risk of end-stage renal disease (ESRD) among these patients remain sparse. We assessed the short-term and long-term risk of ESRD after D-AKI, compared it with the risk in other ICU patients, and examined the risk within subgroups of ICU patients.MethodsWe used population-based medical registries to identify all adult patients admitted to an ICU in Denmark from 2005 through 2010, who survived for 90 days after ICU admission. D-AKI was defined as needing acute dialysis at or after ICU admission. Subsequent ESRD was defined as a need for chronic dialysis for more than 90 days or a kidney transplant. We computed the cumulative ESRD risk for patients with D-AKI and for other ICU patients, taking into account death as a competing risk, and computed hazard ratios (HRs) using a Cox model adjusted for potential confounders.ResultsWe identified 107,937 patients who survived for 90 days after ICU admission. Of these, 3,062 (2.8%) had an episode of D-AKI following ICU admission. The subsequent risk of ESRD up to 180 days after ICU admission was 8.5% for patients with D-AKI, compared with 0.1% for other ICU patients. This corresponds to an adjusted HR of 105.6 (95% confidence interval (CI): 78.1 to 142.9). Among patients who survived 180 days after ICU admission without developing ESRD (n = 103,996), the 181-day to 5-year ESRD risk was 3.8% for patients with D-AKI, compared with 0.3% for other ICU patients, corresponding to an adjusted HR of 6.2 (95% CI: 4.7 to 8.1). During the latter period, the impact of AKI was most pronounced in the youngest patients, aged 15 to 49 years (adjusted HR = 12.8, 95% CI: 6.5 to 25.4) and among patients without preexisting chronic kidney disease (adjusted HR = 11.9, 95% CI: 8.5 to 16.8).ConclusionD-AKI is an important risk factor for ESRD for up to five years after ICU admission.

No abstract

ObjectivesAcute kidney injury (AKI) is a frequent postoperative complication, but the mortality impact within different postoperative time frames and severities of AKI are poorly understood. We examined the occurrence of postoperative AKI among colorectal cancer (CRC) surgery patients and the impact of AKI on mortality during 1 year after surgery.DesignObservational cohort study. We defined the exposure, AKI, as a 50% increase in plasma creatinine or initiation of renal replacement therapy within 7 days after surgery or an absolute increase in creatinine of 26 µmol/L within 48 hours.SettingPopulation-based Danish medical databases.ParticipantsA total of 6580 patients undergoing CRC surgery in Northern Denmark during 2005–2011 were included from the Danish Colorectal Cancer Group database.Outcomes measureOccurrence of AKI and 8–30, 31–90 and 91–365 days mortality in patient with or without AKI.ResultsAKI occurred in 1337 patients (20.3%) of the 6580 patients who underwent CRC surgery. Among patients with AKI, 8–30, 31–90 and 91–365 days mortality rates were 10.1% (95% CI 8.6% to 11.9%), 7.8% (95% CI 6.4% to 9.5%) and 12.0% (95% CI 10.3% to 14.2%), respectively. Compared with patients without AKI, AKI was associated with increased 8–30 days mortality (adjusted HR (aHR)=4.01,95% CI 3.11 to 5.17) and 31–90 days mortality (aHR 2.08,95% CI 1.60 to 2.69), while 91–365 days aHR was 1.12 (95% CI 0.89 to 1.41). We observed no major differences in stratified analyses.ConclusionsAKI after surgery for CRC is a frequent postoperative complication associated with a substantially increased 90-day mortality. AKI should be considered a potential target for reducing 90-day mortality.

No abstract