Low Admission Plasma Creatinine Concentration and In-Hospital MortalityWisit Cheungpasitporn

Disclosure• None

Crit Care Med. 2013;41(8):1878-83.

• Observational cohort study in Dutch 154,308 ICU patients

• Hospital mortality, increasing for BMI< 18.5 kg/m2

• BMI 30–39.9 kg/m2 had the lowest risk of death with an adjusted OR of 0.86 (0.83–0.90)

Background

• BMI suffers from numerous practical limitations in the ICU (such as accurately weighing patients).

• Several studies reported on the association between low BMI and poor outcome of critical illness .

• BMI does not readily discriminate between lean, fat, and water body mass.

Cartin-Ceba R1, Afessa B, Gajic O. Crit Care Med 2007;35:2420-3.

Crit Care Med 2007;35:2420-3.

• A retrospective cohort study of 11,291 adult patients

• Mayo Clinic Hospital—Rochester ICUs

• Between 2003 and 2006

Crit Care Med 2007;35:2420-3.

Crit Care Med. 2016 Jan;44(1):73-82.

Objective• To evaluate the independent association

between low peak admission plasma creatinine concentrations and in-hospital mortality in patients requiring critical care in Australia and New Zealand.

Methods: Study design• Multicenter, binational, retrospective cohort study.

• The Australian and New Zealand Intensive Care Society (ANZICS) Centre for Outcome and Resource Evaluation and adult patient database (APD)

• Data on over 1.5 million admission to 175 ICUs in Australia and New zealand

• 2000 – 2013

• Exclusion criteria:• Readmission episodes, missing in-hospital mortality, post-

kidney transplant, chronic RRT, missing peak plasma creatinine

Patients’ flow chart

Creatinine and Outcome• Peak plasma creatinine

• The highest value in the first 24 hours of ICU admission

• Creatinine (µmol/L) were categorized into < 30, 30-39, 40-49, 50-59, 60-69, 70-79, 80-89, 90-99, 100-119,120-139, 140-179 and ≥ 180

• Reference group: 70-79 µmol/L (0.79-0.89 mg/dL)

• Primary outcome• In-hospital mortality

Statistical analysis• Association with mortality and serum creatinine

• Logistic regression models• Adjusting for a priori-defined list of covariates

• Illness severity• Site of admission• Age• Gender• Year of admission

• 2-sided p value < 0.01 considered statistically significant

Result - Characteristics

1.23 (1.13)

Result

All patient (n=1,047,518)

Adjusted for year of ICU admission, age, gender, institution, ANZROD predicted mortality

No significant interaction between creatinine and year of ICU admission

0.79-0.89<0.34 ≥ 2.04

<0.340.34-0.440.45-0.550.57-0.670.68-0.780.79-0.890.90-1.011.02-1.121.13-1.351.36-1.571.58-2.02≥2.04

Patient with height and weight date (n=96,630)

Adjusted for year of ICU admission, age, gender, institution, weight, height, ANZROD predicted mortality

0.79-0.89<0.34 ≥ 2.04

<0.340.34-0.440.45-0.550.57-0.670.68-0.780.79-0.890.90-1.011.02-1.121.13-1.351.36-1.571.58-2.02≥2.04

Non-operative admission (n=465,771)

Adjusted for year of ICU admission, age, gender, institution, ANZROD predicted mortality

0.79-0.89<0.34 ≥ 2.04

Postoperative admission (n=572,320)

Adjusted for year of ICU admission, age, gender, institution, ANZROD predicted mortality

0.79-0.89<0.34 ≥ 2.04

Elective surgical admission (n=470,594)

Adjusted for year of ICU admission, age, gender, institution, ANZROD predicted mortality

0.79-0.89<0.34 ≥ 2.04

Trauma admission (n=56,450)

Adjusted for year of ICU admission, age, gender, institution, ANZROD predicted mortality

0.79-0.89<0.34 ≥ 2.04

Infection-related admission (n=148,164)

Adjusted for year of ICU admission, age, gender, institution, ANZROD predicted mortality

0.79-0.89<0.34 ≥ 2.04

<0.340.34-0.44

0.45-0.550.57-0.670.68-0.78

<0.340.34-0.440.45-0.550.57-0.670.68-0.78

<0.340.34-0.44

0.45-0.550.57-0.670.68-0.78

<0.340.34-0.440.45-0.550.57-0.670.68-0.78

<0.340.34-0.44

0.45-0.550.57-0.67

0.68-0.78

<0.340.34-0.440.45-0.550.57-0.670.68-0.78

Discussion

• This study demonstrated a significant association between low peak plasma Cr concentrations and increased in-hospital mortality.

• Peak plasma Cr <60 μmol/L (<0.68 mg/dL) in the first 24 hours after ICU admission imply a steadily increasing adjusted mortality risk, compared with a peak plasma Cr of 70–79 μmol/L (0.79-0.89 mg/dL).

Discussion

• In cases where Cr value < 30 μmol/L (<0.34 mg/dL), the adjusted odds of dying in-hospital is > two-fold higher than the reference category and exceeds the risk implied with elevated (≥ 180 μmol/L) values (≥ 2.04 mg/dL).

Thongprayoon C, Cheungpasitporn W, Kashani K. J Thorac Dis. 2016;3(2).

Sime FB et al. Curr Opin Pharmacol. 2015 Oct;24:1-6.

Limitations

• Retrospective cohort study• Information on fluid balance, feeding, functional

status, and physical therapy are lacking, all of which are likely to intimately influence these results.

Conclusion

• In a large heterogenous cohort of critically ill patients, low admission peak plasma creatinine concentrations are independently associated with increased risk-adjusted in-hospital mortality.

• Further research should now focus on the potential mechanisms underpinning this finding, such as a low skeletal muscle mass and/or fluid overload.

Thongprayoon C, Cheungpasitporn W, Kashani K. J Thorac Dis. 2016;3(2).

A Novel Index for Estimation of Muscle Mass Using Biomarkers of Kidney Function: Sarcopenia Index

Kashani K, Kukrálová L, Kashyap R, Frazee E. Am J Respir Crit Care Med 191: A2284, 2015

A Novel Index for Estimation of Muscle Mass Using Biomarkers of Kidney Function: Sarcopenia Index

• Critically ill patients • Mayo Clinic Hospital, Rochester ICU • October 2008-December 2010

*SI = (serum creatinine/serum cystatin C) X 100Am J Respir Crit Care Med 191: A2284, 2015

Low admission serum creatinine concentration predicts mortality in in hospitalized patients independent of body mass index

• Observational cohort study in 73,994 patients

• Mayo Clinic Hospital—Rochester ICUs

• Between 2011 and 2013

Thongprayoon C, Cheungpasitporn W, Kittanamongkolchai W, Harrison AM, Kashani K.Am J Med. 2016 Dec 18. pii: S0002-9343(16)31234-7. doi:

10.1016/j.amjmed.2016.11.020. [Epub ahead of print]PMID: 27998681

In-hospital MortalityOutcome Serum creatinine level at hospital admission (mg/dL)

≤0.4 0.5-0.6 0.7-0.8 0.9-1.0 1.1-1.2 1.3-1.4 ≥1.5

Hospital mortality

25 (2.8) 93 (1.0) 173 (0.8) 187 (1.0) 136 (1.3) 113 (2.4) 246 (3.6)

Mortality, OR (95%CI)

             

Model 1: unadjusted

3.69 (2.35-5.52)

1.23 (0.95-1.58)

1 (ref) 1.24 (1.01-1.53)

1.71 (1.37-2.15)

3.16 (2.48-4.01)

4.81 (3.96-5.86)

Model 2# 3.29 (2.08-5.00)

1.28 (0.99-1.66)

1 (ref) 1.13 (0.91-1.39)

1.29 (1.03-1.64)

1.96 (1.52-2.51)

2.56 (2.07-3.17)

Model 3: Model 2 and BMI

3.40 (2.10-5.28)

1.34 (1.02-1.7)

1 (ref) 1.14 (0.91-1.43)

1.34 (1.04-1.72)

2.10 (1.60-2.74)

2.65 (2.11-3.35)

#model 2: Adjusted for age, sex, race, principal diagnosis, Charlson comorbidity score, CAD, CHF, PVD, stroke, DM, COPD and cirrhosis.

Thongprayoon C, Cheungpasitporn W, Kittanamongkolchai W, Harrison AM, Kashani K.Am J Med. 2016 Dec 18. pii: S0002-9343(16)31234-7. doi:

10.1016/j.amjmed.2016.11.020. [Epub ahead of print]PMID: 27998681

Questions & Discussion