Association between Hyperkalemia at Critical Care Initiation and Mortality
Text for electronic repository of the journal
Gearoid M. McMahon, MD, Renal Division, Brigham and Women's Hospital
Mallika L. Mendu, MD, MBA, Department of Internal Medicine, Brigham and Women’s
Hospital
Fiona K. Gibbons, MD, Pulmonary Division, Massachusetts General Hospital
Kenneth B. Christopher, MD, FASN, FCCP, The Nathan E. Hellman Memorial Laboratory, Renal Division, Brigham and Women's Hospital
Materials and Methods
Data Sources
The RPDR has collected electronic data prospectively on 3 million patients which are stored for research purposes in dedicated data servers by Partners Healthcare the parent corporation that operates the hospitals under study. The following data were retrieved: Demographics, Vital status for up to 10 years following critical care initiation, hospital admission and discharge date, laboratory values, transfusion reports, medications, Diagnosis Related Group (DRG) assigned at discharge, ICD-9CM codes, and Current Procedural Terminology (CPT) codes for in-hospital procedures and services.
The following data were retrieved: Demographics, Vital status for up to 10 years following critical care initiation, hospital admission and discharge date, laboratory values, transfusion reports, medications, Diagnosis Related Group (DRG) assigned at discharge, ICD-9CM codes, and Current Procedural Terminology (CPT) codes for in-hospital procedures and services.
Covariates
HbA1c was measured within 1 year prior and up to 1 week following hospital admission [1] that resulted in critical care, using the HbA1c value closest to hospital admission. Cohort patients were considered to have received medications if the electronic pharmacy data showed a prescription up to 7 days prior to potassium measurement.
Sepsis is defined by the presence of any of the following ICD-9-CM codes: 038.0-038.9, 020.0, 790.7, 117.9, 112.5, or 112.81, 3 days prior to critical care initiation to 7 days after critical care initiation [2]. Acute myocardial infarct (AMI) is defined by ICD-9-CM 410.0-410.9 prior to or on day of critical care initiation [3]. Diabetes mellitus (DM) is defined by ICD-9-CM code 250.xx in outpatient or inpatient records in the prior two years [4-5]. Acute kidney injury (AKI) was defined as ICD-9-CM 584.5, 584.6, 584.7, 584.8, or 584.9 seven days prior to three days after critical care initiation [6]. Number of failed organs was adapted from Martin et al[2] and defined by a combination of ICD-9-CM and CPT codes relating to acute organ dysfunction assigned from 3 days prior to critical care initiation to 30 days after critical care initiation [7-8]. The number of failed organs is the summation of the number of organ categories (Respiratory Failure, Cardiovascular Failure, Renal, Hepatic, Hematologic, Metabolic and or Neurologic)[2] present by ICD-9 code assignment from 3 days prior to critical care initiation to 30 days after critical care initiation. We have shown that the number of failed organs variable is strongly associated with mortality following critical care [9]. Chronic Kidney Disease Stage was determined by the Modification of Diet in Renal Disease (MDRD) equation from baseline creatinine [10]. Patients with end stage renal disease prior to critical care initiation were identified by submitting the administrative data to the United States Renal Data System a national data system that collects, analyzes, and distributes ESRD clinical and claims data to the US Centers for Medicare and Medicaid Services [11].
As hyperkalemia may be associated with transfusion [12], cohort patients who received red blood cell transfusions in the 48 hours prior to potassium measurement were considered to have received transfusions in the analysis. Records of the administration of total parenteral nutrition (TPN) in the 48 hours prior to potassium measurement was determined by CPT code 99.15 and confirmed by pharmacy records. Cohort patients were considered to have received intravenous or oral potassium promoting medications (Table 2) if inpatient medication records listed an order up to 7 days prior to potassium measurement.
Patient Type is defined as Medical or Surgical and incorporates the Diagnostic Related Grouping (DRG) methodology [13]. Procedures were determined by CPT codes as follows: Renal replacement therapy (RRT) seven days prior or three day following critical care initiation (CPT codes 90935, 90945), CABG surgery performed on the day prior or the day after critical care initiation (CPT codes 33510 to 33536).
The Deyo–Charlson index was used to assess the burden of chronic illness [14]. We employed the ICD-9 coding algorithms developed by Quan et al [15] to derive a Deyo–Charlson index for each patient. The algorithms for ICD-9 coding from administrative data have been validated for administrative data [15].
Acute kidney injury was defined in a subset of the cohort that had baseline creatinine measured as RIFLE class Injury or Failure [16] occurring between seven days prior to critical care initiation to the day of critical care initiation. We only applied the serum creatinine criteria to determine the maximum RIFLE class [17]. We classified patients according to the maximum RIFLE class [18] defined as a fold change in serum creatinine from pre-admission serum creatinine [17]. RIFLE class was defined as Risk (fold change ≥ 1.5), Injury (fold change ≥ 2.0) or Failure (fold change ≥ 3.0) [18]. Patients with baseline serum creatinine of >4.0 mg/dl who had an absolute change in serum creatinine > 0.5 mg/dl were defined as RIFLE class Failure [16]. Pre-admission creatinine was obtained in patients from 7 to 365 days prior to hospital admission with the creatinine closet to hospital admission utilized.
Results
Table 3 presents the OR of 30-day mortality for creatinine (per 1 mg/dL) of 0.81 which appears to be counter to published studies on the risk of increased creatinine in the ICU and mortality [19-20]. The odds ratio for creatinine presented in Table 3 is specified in a way that minimizes confounding of the potassium-mortality relationship and was tested empirically. The linear creatinine variable was chosen for the analysis as it performed better than the categorical and was more parsimonious. Further, the OR for creatinine (per 1 mg/dL) in Table 3 is presented as adjusted for all factors listed in the table. To show the creatinine-mortality relationship in our cohort, we analyzed the data with categorical creatinine as the exposure of interest and 30-day mortality as the outcome. The adjusted OR for 30-day mortality is as follows: Cr < 0.8 mg/dl OR 1.26 (95%CI 1.14-1.38; p<.0001), Cr 1.1-1.5 mg/dl OR 1.39 (95%CI 1.28-1.51; p<.0001), Cr 1.5-3.0 mg/dl OR 2.61 (95%CI 2.41-2.84; p<.0001), Cr >3.0 mg/dl OR 2.95 (95%CI 2.67-3.27; p<.0001) all referent to Cr 0.8-1.0 mg/dl. Estimates were adjusted for age, gender, race and Deyo-Charlson Index. The categorical creatinine analysis above is consistent with previous publications regarding increased mortality with increased creatinine in the ICU [19-20]. Further, the OR of 30-day mortality presented in Table 3 in patients with AKI was 1.72 and in patients who require renal replacement was 3.30 which are consistent with prior studies [21-22].
Subanalyses
In a subanalysis of patients with pH present within 4 hours of potassium draw (n=21,500) we substituted pH for HCO3 in the multivariable analysis. The adjusted risk of 30-day mortality was 1.3- and 1.4-fold higher in patients in the K 6.0-6.5 mEq/L and K > 6.5 mEq/L groups, respectively, compared with those in K 4.0-4.5 mEq/L group (K 6.0-6.5 mEq/L OR 1.31; 95% CI, 1.07-1.61; p=0.01; K > 6.5 mEq/L OR 1.37; 95% CI, 1.16-1.62; <.0001). A subanalysis of patients was performed with data present to determine baseline chronic kidney disease stage via the MDRD equation and acute kidney injury defined as RIFLE class Injury and Failure (n=15,228). When baseline chronic kidney disease stage replaced the creatinine variable and RIFLE class Injury and Failure was utilized in the place of ICD-9 defined AKI, the multivariable adjusted risk of 30-day mortality was 1.4- and 1.5-fold higher in patients in the K 6.0-6.5 mEq/L and K > 6.5 mEq/L groups, respectively, compared with those in K 4.0-4.5 mEq/L group (K 6.0-6.5 mEq/L 1.43 95% CI; 1.08-1.89 p=0.01; K > 6.5 mEq/L 1.47 95% CI; 1.18-1.83 p=0.01).
Discussion
The etiology of hyperkalemia is related to increased potassium intake, intracellular redistribution of potassium, decreased potassium excretion or a combination thereof. The clinical importance of potassium is related to the ratio of intracellular (98%) to extracellular potassium which is the principal determinant of the resting membrane potential. Movement of 1-2% of intracellular potassium to the extracellular compartment can increase the serum K by 4.0 mEq/L [23]. Decreased potassium excretion in the setting of renal insufficiency contributes to over 80% of hyperkalemic episodes [24], commonly occurring in combination with potassium supplementation [25-27]. Cellular shift of potassium can be seen in beta-blockade, insulin resistance, inorganic acidemia, digitalis toxicity, hypertonicity, or cell destruction. The dominant mechanism in hyperkalemia is impaired renal K excretion.
While studies outside the ICU have shown a correlation between potassium > 6.0 mEq/L and morbidity [28-30], there is little understanding of the consequence of levels near the normal range [31]. In a study of NHANES I data from a general population sample, serum potassium 4.5–5.4 mEq/L was independently associated with increased cardiovascular mortality [31]. Recently a large study of in-patients with acute myocardial infarction showed that potassium ≥ 4.5 mEq/ml was associated with increased in hospital mortality [32].
We observe a decreased mortality rate for diabetic ICU patients (Table 3). The literature on this point is conflicting, with reports of increased, equal or even decreased mortality rates among the diabetic population compared to patients without diabetes. A large recent meta-analysis of 141 studies showed that diabetes is not associated with increased mortality risk in any ICU population except cardiac surgery patients [33]. Other studies have shown mortality related to critical care is observed to be higher in non-diabetics [34-40].
Electronic Repository References
1. Aguilar D, Bozkurt B, Ramasubbu K, Deswal A, (2009) Relationship of hemoglobin A1C and mortality in heart failure patients with diabetes. J Am Coll Cardiol 54: 422-428
2. Martin GS, Mannino DM, Eaton S, Moss M, (2003) The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med 348: 1546-1554
3. Trespalacios FC, Taylor AJ, Agodoa LY, Abbott KC, (2002) Incident acute coronary syndromes in chronic dialysis patients in the United States. Kidney Int 62: 1799-1805
4. Chen G, Khan N, Walker R, Quan H, (2010) Validating ICD coding algorithms for diabetes mellitus from administrative data. Diabetes Res Clin Pract 89: 189-195
5. Zgibor JC, Orchard TJ, Saul M, Piatt G, Ruppert K, Stewart A, Siminerio LM, (2007) Developing and validating a diabetes database in a large health system. Diabetes Res Clin Pract 75: 313-319
6. Waikar SS, Wald R, Chertow GM, Curhan GC, Winkelmayer WC, Liangos O, Sosa MA, Jaber BL, (2006) Validity of International Classification of Diseases, Ninth Revision, Clinical Modification Codes for Acute Renal Failure. J Am Soc Nephrol 17: 1688-1694
7. Braun A, Chang D, Mahadevappa K, Gibbons FK, Liu Y, Giovannucci E, Christopher KB, (2011) Association of low serum 25-hydroxyvitamin D levels and mortality in the critically ill*. Crit Care Med 39: 671-677
8. Beier K, Eppanapally S, Bazick HS, Chang D, Mahadevappa K, Gibbons FK, Christopher KB, (2011) Elevation of blood urea nitrogen is predictive of long-term mortality in critically ill patients independent of "normal" creatinine. Crit Care Med 39: 305-313
9. Braun AB, Gibbons FK, Litonjua AA, Giovannucci E, Christopher KB, (2012) Low serum 25-hydroxyvitamin D at critical care initiation is associated with increased mortality*. Crit Care Med 40: 63-72
10. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D, (1999) A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 130: 461-470
11. Meduri GU, Golden E, Freire AX, Taylor E, Zaman M, Carson SJ, Gibson M, Umberger R, (2007) Methylprednisolone infusion in early severe ARDS: results of a randomized controlled trial. Chest 131: 954-963
12. de Silva M, Seghatchian MJ, (1994) Is depletion of potassium in blood before transfusion essential? Lancet 344: 136
13. Rapoport J, Gehlbach S, Lemeshow S, Teres D, (1992) Resource utilization among intensive care patients. Managed care vs traditional insurance. Arch Intern Med 152: 2207-2212
14. Charlson ME, Pompei P, Ales KL, MacKenzie CR, (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40: 373-383
15. Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, Saunders LD, Beck CA, Feasby TE, Ghali WA, (2005) Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care 43: 1130-1139
16. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P, (2004) Acute renal failure - definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 8: R204-212
17. Bagshaw SM, Uchino S, Cruz D, Bellomo R, Morimatsu H, Morgera S, Schetz M, Tan I, Bouman C, Macedo E, Gibney N, Tolwani A, Oudemans-van Straaten HM, Ronco C, Kellum JA, (2009) A comparison of observed versus estimated baseline creatinine for determination of RIFLE class in patients with acute kidney injury. Nephrol Dial Transplant 24: 2739-2744
18. Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC, De Bacquer D, Kellum JA, (2006) RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit Care 10: R73
19. Smith GL, Shlipak MG, Havranek EP, Foody JM, Masoudi FA, Rathore SS, Krumholz HM, (2006) Serum urea nitrogen, creatinine, and estimators of renal function: mortality in older patients with cardiovascular disease. Arch Intern Med 166: 1134-1142
20. Hillege HL, Nitsch D, Pfeffer MA, Swedberg K, McMurray JJ, Yusuf S, Granger CB, Michelson EL, Ostergren J, Cornel JH, de Zeeuw D, Pocock S, van Veldhuisen DJ, (2006) Renal function as a predictor of outcome in a broad spectrum of patients with heart failure. Circulation 113: 671-678
21. Metnitz PG, Krenn CG, Steltzer H, Lang T, Ploder J, Lenz K, Le Gall JR, Druml W, (2002) Effect of acute renal failure requiring renal replacement therapy on outcome in critically ill patients. Crit Care Med 30: 2051-2058
22. Uchino S, Kellum JA, Bellomo R, Doig GS, Morimatsu H, Morgera S, Schetz M, Tan I, Bouman C, Macedo E, Gibney N, Tolwani A, Ronco C, (2005) Acute renal failure in critically ill patients: a multinational, multicenter study. JAMA 294: 813-818