sign in sign up
<<
Home , Calcinosis, Hyperkalemia

BRIEF REVIEW www.jasn.org

How Dangerous Is ?

† ‡ John R. Montford* and Stuart Linas*

*Division of Renal Disease and Hypertension, Department of Medicine, University of Colorado Anschutz Medical Campus, Aurora, Colorado; †Renal Section, Medicine Service, Veterans Affairs Eastern Colorado Health System, Denver, Colorado; and ‡Division of , Department of Medicine, Denver Health and Hospitals, Denver, Colorado

ABSTRACT Hyperkalemia is a potentially life-threatening disorder appreciated with lower the resting cardiac membrane po- greater frequency in patients with renal disease, failure, and with use of certain tential. This decreases the threshold for such as renin angiotensin inhibitors. The traditional views rapid phase-0 Na+-dependent depolar- that hyperkalemia can be reliably diagnosed by electrocardiogram and that partic- ization resulting in an increase in cardiac ular levels of hyperkalemia confer cardiotoxic risk have been challenged by several conduction velocity.7 By electrocardio- reports of patients with atypic presentations. Epidemiologic data demonstrate gram (ECG), these changes are manifes- strong associations of morbidity and mortality in patients with hyperkalemia but ted by “peaked” or “tented” Twaves these associations appear disconnected in certain patient populations and in differ- most prominent in the precordial (V2– ing clinical presentations. Physiologic adaptation, structural cardiac disease, medi- V4) leads. With larger acute rises in ex- cation use, and degree of concurrent illness might predispose certain patients tracellular concentration, presenting with hyperkalemia to a lower or higher threshold for toxicity. These conduction delay becomes prominent factors are often overlooked; yet data suggest that the clinical context in which in the atrioventricular node and His– hyperkalemia develops is at least as important as the degree of hyperkalemia is in Purkinje system due to determining patient outcome. This review summarizes the clinical data linking shortening and prolongation of phase-4 hyperkalemia with poor outcomes and discusses how the efficacy of certain treat- diastolic .7,8 Indeed, pro- ments might depend on the clinical presentation. longation of the PR interval, P-wave am- plitude, and increased QRS complex J Am Soc Nephrol 28: 3155–3165, 2017. doi: https://doi.org/10.1681/ASN.2016121344 width are ominous findings in patients with advanced hyperkalemia that can precede a classically described “sine-wave” INTEGRATED DISCUSSION processes. Blunted potassium redistribu- pattern on ECG.9 Thus, hyperkalemia tion typically occurs through predisposes to both cardiac hyperexcit- Introduction deficiency, decreases in aldosterone bio- ability (ventricular tachycardia, ventricular Hyperkalemia is an electrolyte distur- synthesis or action, diminished adrener- fibrillation) and depression (, bance occurring with increased fre- gic signaling, and osmolar disturbances , interventricular quency among patients with CKD, including hyperglycemia. Renal failure, conduction delay, and ), both of diabetes, heart failure, and use of certain and/or failure to augment distal tubular which can be fatal. medicationssuchasreninangiotensin potassium secretion, is largely respon- Despite a wealth of animal data dem- aldosterone system (RAAS) inhibitors sible for the maintenance of hyperkale- onstrating cardiotoxicity from acute and nonsteroidal anti-inflammatory mia. Many studies reproducibly identify hyperkalemia, these presentations are drugs.1–4 Extracellular potassium con- common clinical risk factors that are as- centration is usually kept within a narrow sociated with the development of hyper- physiologic range by redundant and kalemia regardless of the clinical setting Published online ahead of print. Publication date highly efficient homeostatic mechanisms (Table 1). available at www.jasn.org. that simultaneously control internal po- The fatal consequences of rapid in- Correspondence: Dr. John R. Montford, Division of tassium redistribution while regulating creases in extracellular potassium con- Renal Disease and Hypertension, University of Colorado Anschutz Medical Campus, Research 2, net potassium excretion. Hyperkalemia centration have been demonstrated in Box C281, 12700 E. 19th Avenue, Aurora, CO 80045. occurs when rises in extracellular potas- the setting of acute intravenous potas- Email: [email protected] 5,6 sium concentration are accompanied by sium loading in animals. Early rises Copyright © 2017 by the American Society of one, or additive, defects in these two in extracellular potassium concentration Nephrology

J Am Soc Nephrol 28: 3155–3165, 2017 ISSN : 1046-6673/2811-3155 3155 BRIEF REVIEW www.jasn.org

Table 1. Risk factors for the degrees of hyperkalemia and only ap- to 7.32). Furthermore, as compared development of hyperkalemia proached minimal predictive power with survivors with hyperkalemia, Clinical Risk Factor Exposure with potassium levels of 7.2–9.4 meq/ nonsurvivors had higher increases in se- 17 6 Male sex Potassium supplements L. In patients on with hy- rum potassium preceding death (1.1 Non-black Penicillin G perkalemia, ECG-diagnosed 1.3 versus 2.261.5 meq/L change in DM “tenting” did not predict the serum potassium from admission, re- CVD NSAIDs potassium and substantially lost its sen- spectively). Most cases of hyperkalemia CHF ACEi/ARBs sitivity with increasing patient age and developed during hospitalization (in AKI MRAs presence of diabetes.18 Specificity for hy- 60% of the cohort) with a mean admis- CKD b-adrenergic blockers perkalemia and sudden death at follow- sion potassium level of 5.761.5 meq/L Acidosis up was improved with evaluation of the rising to 7.160.7 meq/L after an aver- Urinary obstruction , T:R wave amplitude in these patients, but ageof17daysoffollow-up. , sensitivity was also diminished. In another retrospective review of Hyperkalemia has also been associ- .39,000 patients admitted to the inten- DM, diabetes mellitus; CVD, cardiovascular disease; NSAIDs, nonsteroidal anti-inflammatory ated with a host of nontraditional ECG sive care unit at two teaching hospitals in drugs; CHF, congestive heart failure; ACEi, changes including T wave inversions19 Boston, Massachusetts between 1997 angiotensin-converting enzyme inhibitor; ARB, and pseudonormalizations,20 bundle and 2007, incident hyperkalemia inde- angiotensin receptor blockade. branch,21 bifascicular,22 sinoatrial pendently predicted mortality at the exit,20 andatypicbundlebranch time of critical care initiation.32 This as- overall uncommon in humans. Reports blocks,8 and ST depressions and eleva- sociation was graded, with even minor of acute hyperkalemia precipitating car- tions.15,23,24 There are even reports of elevations in serum potassium (to levels diac arrest typically involve intravenous profound hyperkalemia with minimal 4.5–5.0 meq/L) conferring an increased potassium loading, massive turnover, or no discernable ECG changes.21,25,26 risk of death (OR for death within 30 or shift of potassium in the setting of Additionally, ,27 left days, 1.49; 95% CI, 1.38 to 1.59), and 10–14 surgical anesthesia or critical illness. ventricular hypertrophy,28 early benign remained significant after adjusting for In these cases, the measured potas- repolarization,29 and acute coronary is- many potential confounders prevalent sium concentration was usually normal chemia30 are known to induce T wave in the critical care setting (adjusted OR, shortly before cardiopulmonary arrest; “tenting” in patients with normal serum 1.18; 95% CI, 1.09 to 1.27). Further- and only with rapid increases in serum potassium. Because there are currently more, failure of serum potassium to cor- fi potassium did the ndings of tachy- and inconsistent data supporting the utility rect by .1.0 meq/L within 48 hours after bradyarrhythmias associated with hy- of the ECG in predicting the degree of, initial measurement continued to pre- perkalemia become apparent. These and prognosis with, hyperkalemia, we dict death; whereas, this association extreme situations constitute a small must turn to published data that exam- was attenuated among patients achiev- minority of clinical hyperkalemia in ines the relationship between hyperkale- ing this degree of correction. Khanagavi humans which is often incidental, mia and cardiovascular outcomes. et al.33 reported on hospitalized patients asymptomatic, and of unknown duration. with serum potassium .5.1 meq/L, Additionally, there are many published Hyperkalemia in the Setting of finding that duration of hyperkalemia reports demonstrating a large disconnect Critical Illness and mortality increased substantially between degree of hyperkalemia and ex- Compelling data link hyperkalemia with with concomitant tissue necrosis [haz- pected ECG findings in humans. heightened adverse outcomes in the crit- ard ratio (HR) for death, 4.55; 95% CI, The ECG was observed to be some- ically ill population. In a retrospective 1.74 to 11.90], metabolic acidosis (HR, what unreliable in older studies of pa- analysis of 932 hospitalized adults in 4.84; 95% CI, 1.48 to 15.82), and AKI tients with potassium levels ,6.5 meq/ two Korean medical centers, high rates (HR, 3.89; 95% CI, 1.14 to 13.26). Total L.9,15 Modern studies and case reports of (in 35.2%) and cardiac ar- duration of hyperkalemia was also asso- also support that extreme hyperkalemia rest (in 43.3%) occurred in patients with ciated with death, although the associa- is accompanied by inconsistent findings. serum potassium levels .6.5 meq/L.31 tion was less robust (HR, 1.06; 95% CI, For example, a prospective study exam- Nonsurvivors in this cohort had a higher 1.02 to 1.09). ining treatment strategies for acute hy- prevalence of comorbidities which inde- Data from these studies suggest that perkalemiarevealedonly46%ofall pendently predicted death including not only the absolute level, but the veloc- patients with a serum potassium .6.0 multiorgan failure (odds ratio [OR], ity and duration of hyperkalemia are as- meq/L had ascribable ECG changes.16 7.64; 95% confidence interval [95% sociated withpooroutcomes with critical In another study of hospitalized patients CI], 4.00 to 14.57), malignancy (OR, illness. Although compelling, these stud- with serum potassium levels .6.0 meq/L, 2.88; 95% CI, 1.68 to 4.96), AKI (OR, ies are limited by their retrospective de- the ECG was noted to be completely insen- 2.17; 95% CI, 1.27 to 3.71), and need for sign and weighed down by the severity of sitive at diagnosing mild-to-moderate intensive care (OR, 3.62; 95% CI, 1.79 illness in the subjects. Mortality was high

3156 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3155–3165, 2017 www.jasn.org BRIEF REVIEW

(30.7%) in the cohort by An et al.31 and failure post-AMI also supports that a se- risk of death.43 Potassium increases dur- many patients needed cardiopulmonary rum potassium .5.1 meq/L is associated ing longer intradialytic intervals, and (32%), the majority for with higher risk of death (HR, 2.3; 95% many have attempted to link these fluc- reasons unrelated to hyperkalemia. Al- CI, 1.4 to 3.6).39 tuations to the higher incidence of sud- though initial and sustained hyperkale- The studies by Goyal et al.37 and den cardiac death in patients with ESRD. mia predicted mortality in the study by Grodzinsky et al.38 are strengthened by In a 3-year study of community dwelling McMahon et al.,32 there are no data re- use of robust adjustment models that patients on hemodialysis, the presence of garding the disease-specificclinicalim- control for baseline risk, medications, hyperkalemia (defined as three or more provement or lack of improvement in PCI use, and other pertinent factors. averaged serum potassium levels .6.0 the patients who suffered in-hospital Nevertheless, residual confounding in meq/L over a 6-month period) was one mortality. Furthermore, these studies very ill patients is common in retrospec- of the strongest single predictors of sud- were unable to adequately control for tive analyses, and outcomes in patients den death (HR, 2.7; 95% CI, 1.3 to the severity of patient illness due to a with serum potassium .4.5 meq/L in 5.9).44 lack of physiologic data. the study by Goyal et al.37 were ultimately A recently published retrospective driven by a small number of individuals observational trial of 52,734 patients Hyperkalemia during Acute (11%, 2%, and 0.6% of the entire cohort on a Monday/Wednesday/Friday hemo- had potassium levels 4.5–5.0, 5.1–5.5, schedule revealed that serum Hyperkalemia was not originally identi- and .5.5 meq/L, respectively). These potassium levels 5.5–6.0 meq/L were as- fied as a potential risk factor for poor patients also had substantially higher co- sociated with higher risk for subsequent outcomes during evolution of acute morbidities, and lower rates of PCI, as- hospitalization, emergency department myocardial infarction (AMI),34 despite pirin, RAAS inhibitor, b-blocker, and visits, and mortality within 4 days of the existence of strong biologic plausi- statin usage. The analysis by Krogager measurement.45 Of note, the association bility in animal models.35,36 However, et al.39 suffers a similar limitation, in- between hyperkalemia and hospitaliza- modern approaches, including percuta- cluding very few patients with serum po- tion was magnified among patients neous coronary intervention (PCI) and tassium levels .5.1 meq/L. In the study entering a longer intradialytic interval more widespread adoption of RAAS in- by Goyal et al.37 there was also a para- (adjusted OR for hospitalization, 1.12; hibitors, b-adrenergic blockers, and doxic dissociation between rates of ma- 95% CI, 1.0 to 1.24; OR, 1.04; 95% CI, receptor antagonists lignant and rates of overall 0.94 to 1.16; and OR, 1.68; 95% CI, 1.22 (MRAs), have drastically improved pa- mortality in higher versus lower ranges to 2.30 for patients with potassium mea- tient survival while also predisposing of hyperkalemia and normokalaemia, in surements performed on Monday, the post-AMI population to more fre- which both associations were more con- Wednesday, and Friday, respectively). quent hyperkalemia. One widely cited gruent. The authors hypothesized that The association of mortality with hy- retrospective trial of 38,689 hospitalized poor coding of arrhythmias associated perkalemia also appears to extend to pa- patients with AMI treated in the modern with extremes of hyperkalemia, such as tients with earlier stage CKD. Einhorn era demonstrated an independent in- sinus arrest and asystole, led to this et al.2 conducted a retrospective analysis crease in mortality among patients with discrepancy. of 245,808 United States adult veterans potassium levels .5.1 meq/L (OR, 3.27; with and without CKD, showing potas- 95% CI, 2.52 to 4.24) which persisted in Hyperkalemia with CKD sium levels of .5.5 meq/L predicted patients with serum potassium levels of One of the first studies to demonstrate an death just 1 day after measurement. 4.5–5.0 meq/L (OR, 1.99; 95% CI, 1.68 independent association of hyperkale- In another study of 36,000 patients in to 2.36).37 A subsequent analysis of this mia and risk of subsequent death the Cleveland Clinic system with same cohort showed elevated in-hospital involved a large retrospective study of an eGFR,60 ml/min per 1.73 m2,sus- mortality with exposure to a higher Japanese patients with advanced CKD tained hyperkalemia (defined as an aver- number of hyperkalemic episodes presenting for dialysis initiation.40 An age serum potassium level .5.5 meq/L (13.4%, 16.2%, and 19.8% increase in initial serum potassium level .5.5 over 2.3 years) was also associated with mortality with one, two, and three or meq/L at dialysis vintage was the stron- increased all-cause mortality.46 more potassium measurements .5.0 gest single independent predictor of However, an interesting paradox is meq/L, respectively) and maximum mortality after an average of 15 years of documented in these later studies regard- achieved serum potassium level (4.2%, follow-up. In patients on hemodialysis, ing the relationship between CKD stage, 11.1%, 16.6%, 26.6%, and 31.7% in- potassium levels .5.641 and .5.742 hyperkalemia, and mortality. In the study crease in mortality with potassium levels meq/L have been associated with higher by Einhorn et al.2 the strongest associa- ,5.0, 5.0–5.5, 5.5–6.0, 6.0–6.5, and mortality. This is also reflected in pa- tion between hyperkalemia and 1-day .6.5 meq/L, respectively).38 Another tients on peritoneal dialysis, with one mortality involved patients with normal retrospective trial analyzing 90-day mor- study suggesting hyperkalemia .5.5 renal function (OR, 10.32 and 31.64 for tality in 2596 Danish patients with heart meq/L is associated with a heightened serum potassium ranges $5.5 and ,6.0

J Am Soc Nephrol 28: 3155–3165, 2017 How Dangerous Is Hyperkalemia? 3157 BRIEF REVIEW www.jasn.org and $6.0 meq/L, respectively), and de- CKD population and could lower ar- generalizations that can be made from clined as CKD stage progressed; with rhythmogenic potential with concurrent this study. stage 5 CKD associated with a much hyperkalemia. All of these factors could There is conflicting data regarding the lower relative risk (OR, 2.31 and 8.02 be further compounded by changes in outcome of patients with hyperkalemia for serum potassium $5.5 and ,6.0 individual solutes, rapid osmolar shifts, exposed to MRAs. Post hoc data68 from and $6.0 meq/L, respectively). Patients high ultrafiltration rates, and myocardial the Post–Acute Myocardial with ESRD in the study by Nakhoul stunning during dialysis treatments. Infarction Heart Failure Efficacy and et al.46 also seemed to be protected Thus, persons with CKD might be Survival Study (EPHESUS) trial indi- with hyperkalemia relative to the entire uniquely predisposed or uniquely pro- cates that eplerenone maintains a mor- cohort (adjusted HR for death, 1.20; tected from cardiotoxicity with hyperka- tality benefit in patients with CKD and 95% CI, 0.91 to 1.58 versus 1.65; 95% lemia relative to other populations, and eGFR,60 ml/min per 1.73 m2 while si- CI, 1.48 to 1.84, respectively). Further- further studies should be performed multaneously predisposing these pa- more, An et al.31 showed a graded de- with these apparent inconsistencies in tients to higher rates of hyperkalemia. crease in risk of death among patients mind. Unfortunately, patients with more ad- with extreme levels of hyperkalemia vanced CKD (serum .2.5 (.6.5 meq/L) as CKD stage increased Hyperkalemia with Selected mg/dl) were excluded from both the (OR for death with stage 2, 3, 4, and 5 Medication Exposure original EPHESUS trial69 and the earlier CKD, 0.52; 95% CI, 0.35 to 0.78, 0.31; Hyperkalemia that develops while ex- Randomized Aldactone Evaluation 95% CI, 0.21 to 0.46, 0.13; 95% CI, posed to certain medications could alter Study.70 It is important to note that no 0.06 to 0.26, and 0.17; 95% CI, 0.11 to the threshold of cardiac toxicity. Cases of hyperkalemia-associated deaths were re- 0.27). Similar results were observed digoxin poisoning have illustrated quite ported in either of these trials. However, among dialysis versus non-CKD patients dramatic rises in serum potassium with in an analysis of the Eplerenone in Mild with hyperkalemia and AMI in the studies associated arrhythmias.64 In the studies Patients Hospitalization and Survival performed by Goyal et al.37 and Grodzinsky performed by Khanagavi et al.33 and Study in Heart Failure, patients with se- et al.38 One prospective observational McMahon et al.32,butnotAnet al.,31 rum potassium levels .5.5 meq/L had a analysis of sustained hyperkalemia and potassium supplementation with hyper- higher risk of all-cause mortality.71 MRA outcomes in patients with creatinine kalemia was linked to heightened exposure was also associated with more clearance ,50 ml/min demonstrated mortality. With regard to the risks of hyperkalemia (.5.0 meq/L) and higher that hyperkalemia in the ranges of 5.0– hyperkalemia while exposed to RAAS in- mortality in a study of 15,803 United 6.0 meq/L (using an average of six mea- hibitors and b-adrenergic blockers, pa- States veterans with established cardio- surements per patient) appeared to be tients in the study by An et al.31 had vascular disease (OR for death, 1.50; well tolerated.47 lower observed mortality. Conversely, 95% CI, 0.40 to 5.64).72 Patients with Adaptive increases in circulating cat- critically ill patients in the study by eGFR,60 ml/min per 1.73 m2 in this echolamines, aldosterone, and augmen- McMahon et al.32 were not afforded study had even worse outcomes while tation of renal and gastrointestinal (GI) similar protection with either agent. Un- on MRAs (OR for death, 1.74; 95% CI, potassium elimination are thought to fortunately, there is no data on patient 1.11 to 2.71). Another recent study ex- blunt hyperkalemia development in mortality stratified by exposure versus amining use in 27,213 CKD and could partially explain this ap- nonexposure to RAAS inhibitors and b predialysis patients in Taiwan demon- parent disconnect in mortality relative to blockers in the studies performed by strated that exposure was independently non-CKD patients.48–53 However, phys- Goyal et al.,37 Grodzinsky et al.,38 and associated with increases in hospitaliza- iologic adaptation incompletely explains Krogager et al.39 Patients with higher tions for heart failure (adjusted HR, why mortality risk could be diminished mortality in these studies were less likely 1.35; 95% CI, 1.08 to 1.67), infection- once hyperkalemia has already been es- to be exposed to either agent. Other data related deaths (adjusted HR, 1.42; 95% tablished. Furthermore, patients with indicate that RAAS inhibitor use is CI, 1.16 to 1.73), and all-cause mortality CKD might be uniquely predisposed to linked with more profound hyperkale- (adjusted HR, 1.35; 95% CI, 1.24 to more not less toxicity with hyperkalemia mia and death in the elderly65 and in 1.46).73 due to a higher prevalence of metabolic patients with diabetic versus nondiabetic derangements (e.g., , aci- CKD.66 In the randomized controlled Implications of Data Supporting an dosis, and elevated uremic solutes) and Hypertension in Hemodialysis Patients Association between Hyperkalemia structural heart disease. Left ventricular Treated with Atenolol or tri- and Mortality hypertrophy,54,55 atrial fibrillation,56,57 al,67 lisinopril predisposed patients To date, the published studies demon- variability,58 heart failure,59 on dialysis to more hyperkalemia strating an association of mortality silent myocardial infarction,60 QT ab- and higher cardiovascular morbidity with hyperkalemia are largely limited normalities,61 and pulmonary hyperten- than atenolol; although, the lack of a to retrospective analyses that do not pro- sion62,63 are all highly prevalent in the control group is a limitation to further vide evidence of causation. Much of the

3158 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3155–3165, 2017 www.jasn.org BRIEF REVIEW published data are also cross-sectional in although clinician adherence was low. efficacy of these drugs in patients with nature, potentially raising more ques- We conclude that prospectively designed lower renal function. Indeed, absence tions than answers. Furthermore, there randomized trials of treatment for hy- of usage is associated with the are sparse data to suggest that treatment perkalemia should be performed with development of hyperkalemia in at-risk of hyperkalemia modifies risk. In valid endpoints (e.g., target potassium patients.47 Patients with CKD and pa- the study of critically ill patients by levels) and outcome measures (e.g., tients who experience diuretic braking, McMahon et al.,32 theriskofmortality mortality, arrhythmia, and health care in which response to a diuretic becomes with hyperkalemia was attenuated in pa- utilization) in mind. Awell designed trial blunted over time, are known to need tients achieving a .1.0 meq/L decrease using accepted therapies to lower serum higher diuretic doses, diuretic rotation, in serum potassium within 48 hours; al- potassium in noncritically ill monitored or combination to maintain a though, it is unclear if this represented hospitalized patients with hyperkalemia therapeutic effect.75 Therapies directed an actual treatment effect. Other studies would be the safest initial study to per- at augmenting GI potassium excretion suggesting that longer durations and form. Presently, we will review the com- have been in use for many years, princi- failure to reverse hyperkalemia are asso- mon treatments for hyperkalemia and the pally with polystyrene sulfate ciated with mortality suffer from similar potential pitfalls in using these agents. (SPS). Generally, SPS has been shown limitations. Of note, An et al.31 attemp- to be unreliable in the acute setting al- ted to control for the effect of treatment Treatment of Hyperkalemia though data on chronic management by giving individual therapies a weighted Many reviews have been published on might support its use.76 Newer agents, score which was plotted against patient this topic and we will only brieflyhigh- such as sodium cyclosilicate survival. Although increasing number of light these strategies. Acute treatment of (ZS-9; AstraZeneca) and the recently targeted interventions for hyperkalemia life-threatening hyperkalemia necessi- Food and Drug Administration (FDA)– was associated with improved patient tates infusion of intravenous to approved (Veltassa; Relypsa), survival, there was no control for other protect against malignant cardiac hyper- have been demonstrated to effectively treatments that were directed at revers- excitability followed by agents which lower serum potassium when administered ing the underlying illness. It is interest- have been proven in humans to rapidly in patients with chronic hyperkalemia at ing to note that when hemodialysis or and effectively shift potassium into levels ,6.5 meq/L.77–80 Furthermore, se- continuous renal replacement therapy the intracellular space. Insulin appears rum potassium may be rapidly lowered were included as treatments for hyper- the most well studied treatment in this within hours by both ZS-979 and pa- kalemia in this study, the improvement regard and its rapid action to shift potas- tiromer,81 suggesting a previously unrecog- in patient mortality was eliminated. Ex- sium is not dependent on receptor- nized role of the upper GI tract in potassium tracorporeal elimination of potassium is ligand signaling and downstream protein regulation. the most efficient and definitive therapy synthesis. Unlike b-adrenergic agonist Correction of by for life-threatening hyperkalemia. and bicarbonate therapy, insulin does mineralocorticoid administration may Therefore, it is surprising that these thera- not lose its efficacy, and might be en- be an effective therapy to reverse hyper- pies were not associated with improvement hanced, in the presence of renal fail- kalemia. Data indicate that urinary losses in mortality whereas others (withdraw of ure.49 Intravenous dextrose is usually of potassium only partially explain the offending medicines, intravenous cal- given to prevent and fur- treatment effect, suggesting a role for en- cium, insulin/dextrose, etc.) were, espe- ther stimulates endogenous insulin hanced intracellular redistribution or cially if hyperkalemia is presumed to be production. Studies suggest that oral augmented GI excretion.82,83 This ther- the proximate cause of mortality. loading may also be an effective apy has enjoyed some recent resurgence, Abrupt incidence and more rapid ve- strategy to increase insulin and reduce with several reports demonstrating suc- locity of hyperkalemia are salient features serum potassium in patients on hemo- cessful treatment of hyperkalemia due in studies such as those performed by An dialysis.74 There is conflicting data re- to a range of causes.84–86 However, small et al.,31 Goyal et al.,37 and Grodzinsky garding the efficacy of b-adrenergic randomized placebo-controlled trials et al.38 More rapid development of hy- agonists and to re- using oral fludrocortisone in patients perkalemia is potentially more cardio- liably shift potassium; however, we ob- on hemodialysis with hyperkalemia toxic, and directed treatment might serve that these agents are often used in have demonstrated poor efficacy87 or have more protective effects in this pop- the acute management of hyperkalemia. modest efficacy88 at lowering serum po- ulation compared with others. Never- Ultimately, acute hemodialysis may be tassium. It is important to note that the theless, we can find only one published necessary for the extracorporeal elimi- doses used in these trials were relatively attempt at protocolizing treatment of nation of potassium in life-threatening low (0.1 mg fludrocortisone daily), and patients who develop hyperkalemia situations. older data indicate that patients with re- while hospitalized.16 This study demon- Diuretics are often underappreciated nal disease require much higher doses strated no significant changes in patient as an effective treatment of hyperkalemia (up to 1.0 mg daily) to effectively reverse outcome with prescribed protocols, owing to misconceptions regarding the hyperkalemia.82

J Am Soc Nephrol 28: 3155–3165, 2017 How Dangerous Is Hyperkalemia? 3159 BRIEF REVIEW www.jasn.org

Perhaps the most underused of all nephrolithiasis, and gout flares. SPS has Global Outcomes nor the Disease therapies to combat hyperkalemia in- been linked with numerous cases of in- Outcomes Quality Initiative have pub- volves reduction of dietary potassium testinal necrosis,97–99 which has given lished guidelines in the treatment of hy- intake. Careful screening of the diet for this drug a warning label by the FDA perkalemia. The Investigator Network potassium-rich foods is not often per- and limited its modern appeal. Newer Initiative Cardiovascular and Renal Clin- formed due to time-crunched clinician potassium exchange resins are not with- ical Trialists recently published guide- visits and poor dietary education given to out potential side effects and have been lines104 on workup of hyperkalemia health care providers. A careful review of shown to induce hypomagnesemia,78 and treatment strategies in patients potassium intake and directed counsel- hypercalciuria,100 and even edema80 at with serum potassium .5.1 meq/L, but ing might prevent incident hyperkalemia high doses. Long-term effects from these doesn’t stipulate exactly who should be and serve as an important adjunct with drugs are unknown and neither of these treated. other therapies in the treatment of hyper- newer agents have been shown to be Onthebasisofthepublisheddata kalemia. However, we can find no human efficacious in patients on dialysis. Exog- demonstrating disparate risks of hyper- data that dietarycounseling is an effective enous are not com- kalemia in different patient populations, strategy in the prevention or treatment of monly used for hyperkalemia given the safety profile and reliability of agents hyperkalemia. Data on the effectiveness concerns for precipitating volume over- to reduce serum potassium, and our own of dietary potassium reduction in hyper- load and significant cardiopulmonary experience, we propose a step-wise strat- kalemic individuals with advanced CKD complications.82 egy for the prevention and treatment of (who often have relatively fixed levels of Central line insertion for acute dialy- hyperkalemia in the following sections urinary and GI potassium excretion) is sis access can predispose to a host of per- which is applicable in a range of clinical particularly needed. iprocedural complications and trauma to settings. We propose treatment on the the central veins that vitally feed future basis of clinical presentation of the pa- Adverse Events Linked to dialysis access creation. Very low potas- tient rather than degree of hyperkalemia, Potassium-Lowering Therapies sium–containing dialysate solutions which poorly predicts cardiotoxicity in Clinicians employ these therapies to pa- (,2.0 meq/L) are sometimes employed humans. Although we support a thresh- tients with hyperkalemia, but to what for severe cases of hyperkalemia but the old for initiating therapy in certain pa- target level? And at what cost to the pa- consequences of achieving a rapid tients, we do not support that any upper tient and the health care system? It is (within minutes) reduction in extracellu- limit of hyperkalemia constitutes an important to highlight that many of the lar potassium concentration are unknown. “emergency” on the basis of the serum treatments used in management of hy- Data suggest that lower potassium–con- potassium concentration alone. Further- perkalemia may have untoward or even taining dialysates are associated with sig- more, we do not recommend guiding unrecognized side effects. Acute infu- nificant morbidity and mortality101–103; therapies on the basis of ECG findings sions of elemental calcium can induce which has largely led to abandoning this given their inherent variability. We heart block in patients with digoxin- practice. Finally, unnecessary hospitali- have found that relying on the ECG with- induced hyperkalemia89 and precipitate zations and clinician hypervigilance out extensive knowledge of the patient’s acute dermal calcifications.90,91 Given may predispose an already frail patient prior cardiac history, velocity of hyper- the higher prevalence of hyperphospha- population to a cascade of unpredictable kalemia development, and baseline temia in patients with CKD, intravenous iatrogenic effects. ECG (all are almost never present) can calcium infusion carries the theoretic distract from addressing the underlying risk of creating or worsening existing cause and interfere with appropriate tar- metastatic vascular calcifications. WHO IS MOST LIKELY TO BENEFIT geted therapy. Hypoglycemia and tachycardia can FROM THE CORRECTION OF accompany insulin and albuterol admin- HYPERKALEMIA AND HOW TO Prevention and Supportive istration, respectively. Large intravenous ACHIEVE IT? AN OPINION-BASED Treatment of Hyperkalemia infusions of sodium bicarbonate may SET OF RECOMMENDATIONS Toreduce the incidence of hyperkalemia, precipitate acute hyperosmolarity,92 in- at risk patients, as defined in Table 1, cluding case reports of central pontine Despite decades of knowledge regarding should be identified and managed in an myelinolysis.93 Sodium bicarbonate in- the potential risks of hyperkalemia, the anticipatory fashion, with dietary mod- fusions also risk development of acute high incidence and prevalence of hyper- ifications, avoidance of medications pulmonary edema,94 ionized hypocalce- kalemia in patients with certain comor- which might worsen risk for hyperkale- mia,94,95 and worsening of AKI and mor- bidities and medication exposures, and mia, and surveillance for common clin- tality in patients undergoing cardiac the availability of effective potassium- ical scenarios that create additive risk. surgery.96 lowering therapies, there are no guide- We advise all patients with existing hy- Diuretics can induce volume contrac- lines to advise who should be treated. perkalemia (.5.0 meq/L) to reduce po- tion, dysnatremias, hypomagnesemia, Neither the Kidney Disease: Improving tassium intake to ,40 meq/d. Similar

3160 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3155–3165, 2017 www.jasn.org BRIEF REVIEW

Table 2. Strategies to urgently treat hyperkalemia Steps Clinical Question Strategy 1. Increase urinary Is the patient volume contracted or euvolemic? Yes, administer trial of volume expansion with or without potassium losses Is the patient volume overloaded or hypertensive? Yes, stratify and treat: a. eGFR.60 ml/min per 1.73 m2 and diuretic-naïve Start low dose loop or -like diuretic b. eGFR,60 ml/min per 1.73 m2 and diuretic-naïve Start moderate dose loop diuretic c. currently taking diuretics Double existing diuretic dose and/or add loop diuretic, thiazide-like diuretic, or carbonic anhydrase inhibitor 2. Increase Does the patient have a contraindication (recent abdominal No, consider a limited trial of patiromer, ZS-9, or SPS gastrointestinal surgery, ileus, obstipation, history of ischemic bowel) to potassium cathartics? elimination 3. Mineralocorticoid Does the patient have a contraindication (greater than stage 1 No, consider a trial of fludrocortisone 0.1 mg daily 3 3–5d replacement HTN, volume overload, history of heart failure) to (In patients with moderately advanced CKD consider mineralocorticoid administration? maintaining or increasing diuretics in tandem) 4. Dialysis Is the patient currently on maintenance dialysis? Yes, optimize dialysis delivery: optimization or Assess delivered dialysis dose, duration, and frequency initiation Screen for patient noncompliance with dialysis and patient/caregiver burnout Address any access dysfunction including poor blood flows, recirculation 2 Assess dialysate K+ and HCO3 concentrations No, revisit steps 1–3 and consider hospitalization and urgent dialysis initiation if hyperkalemia persists SPS, sodium polystyrene sulfate; HTN, hypertension. dietary reductions are advised in patients clinical presentation, is to maintain these incident hyperkalemia with blood prod- with eGFR,30 ml/min per 1.73 m2 and drugs unless other supportive measures uct administration, sepsis, multiorgan in patients with eGFR.30 ml/min per fail to correct the hyperkalemia. failure, myonecrosis, and rewarming 1.73 m2 but prone to hyperkalemia. In We advise cautious administration of of a cooled patient. Patients on dialysis at-risk patients with high-to-normal lev- higher RAAS inhibitor doses and MRAs should have access interventions and els of serum potassium (4.5–5.0 meq/L), in patients with diabetic CKD, advanced other operations scheduled away from proactive dietary screening is advised to CKD, and those with a prior history of long dialytic intervals to minimize the identify and mitigate large potassium hyperkalemia. Combination RAAS in- periprocedural risk of hyperkalemia. Fur- loads. We readily admit these dietary hibitor regimens should be avoided be- thermore, monitoring postprocedural se- recommendations are not based on ev- cause these therapies place patients at rum potassium in patients on dialysis idence supporting efficacy, and they special risk for hyperkalemia without with higher prevalent hyperkalemia is could have the adverse effect of steering proven benefit. Hyperkalemia which de- advised. patients with CKD away from more velops on a diuretic should prompt an nutrient-rich foods. investigation for factors which might Emergency Treatment of Providers should seek to limit or ab- causediureticbrakingandlimitdistal Hyperkalemia stain from exposing higher-risk patients nephron sodium delivery, and thus po- A “Hyperkalemia Emergency,” which we to medications listed in Table 1. In cases tassium secretion. define as a serum potassium .6.0 meq/L where avoidance of these agents is not Hyperkalemia out of proportion to or a sudden increase in serum potassium possible, close monitoring and frequent changes in eGFR should prompt a rigorous 1.0 meq/L above 4.5 meq/L within 24 laboratory checks are advised. The data investigation for urinary obstruction, insu- hours associated with cardiopulmonary on avoidance of b-adrenergic blockers in linopenia, acidosis, and disorders which arrest, evolving critical illness, AMI, or patients prone to hyperkalemia is con- predispose to hypoladosteronism, such as of neuromuscular troversial, because b blocker usage is one adrenal insufficiency. Patients with ad- weakness, should be treated with agents of the few established therapies in CKD vanced CKD, including those on dialysis, that rapidly and reliably shift serum po- and non-CKD patients which is associ- who newly develop hyperkalemia should tassium into the intracellular space while ated with lower risk of cardiovascular be evaluated for new constipation or bowel preparations are made for elimination of events. Our own practice, depending obstruction. In hospitalized patients, the total body potassium (TBK+). Infusion onthedegreeofhyperkalemiaand clinician should be attentive to the risk of of intravenous calcium, insulin, and

J Am Soc Nephrol 28: 3155–3165, 2017 How Dangerous Is Hyperkalemia? 3161 BRIEF REVIEW www.jasn.org dextrose, and lastly inhaled or intrave- higher-risk patients without the need significance in . Arch – nous b-adrenergic agonist therapy to employ intravenous calcium, insulin, Intern Med 169: 1156 1162, 2009 b 3. Weir MR, Rolfe M: Potassium should be administered only in these ex- dextrose, -adrenergic agonists, and bi- and renin-angiotensin-aldosterone system treme circumstances. Consideration for carbonatethathavelittleuseoutside inhibitors. Clin J Am Soc Nephrol 5: 531– sodium bicarbonate administration emergencies and are wholly ineffective 548, 2010 should be given in cases of hyperkalemia at removing TBK+. The inclusion of 4. Lafrance JP, Miller DR: Dispensed selective fl accompanying severe metabolic acido- mineralocorticoid administration is and nonselective nonsteroidal anti-in ammatory drugs and the risk of moderate to severe sis. Arrangements for emergent he- likely a controversial position. In our hyperkalemia: A nested case-control study. modialysis should be made early and own practice, we have observed that Am J Kidney Dis 60: 82–89, 2012 proactively unless a rapidly reversible this strategy is safe and very effective 5. Hiatt N, Katayanagi T, Miller A: Cardiac cause is identified and therapy is expe- in a time-limited fashion in appropri- sensitivity to hyperkalemia in adrenalecto- dited. Our practice might require insti- ately selected patients. Although we mized dogs. Proc Soc Exp Biol Med 149: 168–171, 1975 tuting hemodialysis in patients who include dialysis initiation in cases of re- 6. Coulter DB, Swenson MJ: Effects of potas- might otherwise recover from hyperka- fractory hyperkalemia, outside patients sium intoxication on porcine electrocardio- lemia with directed treatment, yet in this with oligoanuric renal failure and critical grams. Am J Vet Res 31: 2001–2011, 1970 demographic the failure to rapidly stabi- illness we have found there are very few 7. Fisch C: Relation of electrolyte disturbances – lize or reverse hyperkalemia might have circumstances that will require the initi- to cardiac arrhythmias. Circulation 47: 408 419, 1973 fatal consequences. ation of dialysis for hyperkalemia alone. 8. El-Sherif N, Turitto G: Electrolyte disorders We propose that a close follow-up plan and arrhythmogenesis. Cardiol J 18: 233– Urgent Treatment for Hyperkalemia should be implemented for every patient 245, 2011 We propose a more guarded strategy to with hyperkalemia, particularly in the 9. Surawicz B: Relationship between electro- treat hyperkalemia in less dire circum- outpatient setting. Adopting this strat- cardiogram and . Am Heart J 73: 814–834, 1967 stances, which constitutes the majority egy might potentially spare patients 10. Lee SH, Kim KJ, Chang CH, Heo SB: Cardiac of hyperkalemia cases in practice. In pa- from unnecessary emergency depart- arrest from acute hyperkalemia during liver tients with hyperkalemia, but outside ex- ment visits and hospitalizations. Fur- surgery -A case report-. Korean J Anes- tremis, we propose adopting a four-step thermore, employing a strategy such as thesiol 59[Suppl]: S124–S127, 2010 system aimed at maximizing elimination this might better integrate knowledge of 11. Mercer CW, Logic JR: due to + hyperkalemia following intravenous peni- of TBK in a prompt but safe manner: potassium homeostasis by the treating cillin administration. Chest 64: 358–359, (1) Increase urinary potassium clear- clinician and steer both low- and 1973 ance, (2) augment potassium excretion higher-risk patients with hyperkalemia 12. Wiltrout C: Hyperkalemia, bradycardia, and through the GI tract, (3) administer ex- into appropriate, effective, and safe cardiac arrest during percutaneous declot- ogenous mineralocorticoids in selected treatment plans. ting of an arteriovenous graft. Semin Inter- vent Radiol 27: 241–244, 2010 patients, and (4) initiate dialysis or opti- 13. Huggins RM, Kennedy WK, Melroy MJ, mize dialysis delivery (Table 2). We pro- Tollerton DG: Cardiac arrest from pose executing this strategy in patients ACKNOWLEDGMENTS succinylcholine-induced hyperkalemia. Am with normal renal function with serum J Health Syst Pharm 60: 694–697, 2003 potassium .5.0 meq/L, patients with None. 14. Wilson D, Stewart A, Szwed J, Einhorn LH: Cardiac arrest due to hyperkalemia follow- non-ESRD CKD and serum potassium . ing therapy for acute lymphoblastic leuke- 5.5 meq/L, and patients with ESRD mia. Cancer 39: 2290–2293, 1977 on maintenance dialysis and serum po- DISCLOSURES 15. Tarail R: Relation of abnormalities in con- tassium .6.0 meq/L who have failed S.L. reports serving on an advisory board for ZS centration of serum potassium to electro- supportive measures to reverse the hy- Pharma. No financial support was received in the cardiographic disturbances. Am J Med 5: 828–837, 1948 preparation of this manuscript. perkalemia (or are deemed more at risk 16. Acker CG, Johnson JP, Palevsky PM, to fail with supportive measures alone). Greenberg A: Hyperkalemia in hospitalized Utilizing one of these strategies along patients: Causes, adequacy of treatment, with supportive measures should be suf- REFERENCES and results of an attempt to improve physician ficient to reverse most cases of hyperka- compliance with published therapy guide- 1. Sarafidis PA, Blacklock R, Wood E, Rumjon lines. Arch Intern Med 158: 917–924, 1998 lemia; although, several strategies in A, Simmonds S, Fletcher-Rogers J, 17. Montague BT, Ouellette JR, Buller GK: combination can be used to maximize Ariyanayagam R, Al-Yassin A, Sharpe C, Retrospective review of the frequency of efficacy in patients with suspicion of Vinen K: Prevalence and factors associated ECG changes in hyperkalemia. Clin J Am multifactorial hyperkalemia. with hyperkalemia in predialysis patients Soc Nephrol 3: 324–330, 2008 The goalsof this treatment strategy are followed in a low-clearance clinic. Clin J Am 18. Green D, Green HD, New DI, Kalra PA: The Soc Nephrol 7: 1234–1241, 2012 clinical significance of hyperkalaemia- to stabilize and gradually reduce serum 2. Einhorn LM, Zhan M, Hsu VD, Walker LD, associated repolarization abnormalities in potassium in low-risk patients, and more Moen MF, Seliger SL, Weir MR, Fink JC: end-stage renal disease. Nephrol Dial rapidly reduce serum potassium in The frequency of hyperkalemia and its Transplant 28: 99–105, 2013

3162 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3155–3165, 2017 www.jasn.org BRIEF REVIEW

19. Khattak HK, Khalid S, Manzoor K, Stein PK: 35. Hill JL, Gettes LS: Effect of acute coronary and mortality in chronic kidney disease. Am Recurrent life-threatening hyperkalemia with- artery occlusion on local myocardial extra- JNephrol41: 456–463, 2015 out typical electrocardiographic changes. J cellular K+ activity in swine. Circulation 61: 47. Korgaonkar S, Tilea A, Gillespie BW, Kiser Electrocardiol 47: 95–97, 2014 768–778, 1980 M, Eisele G, Finkelstein F, Kotanko P, Pitt B, 20. Yu AS: Atypical electrocardiographic 36. Coronel R, Fiolet JW, Wilms-Schopman FJ, Saran R: Serum potassium and outcomes in changes in severe hyperkalemia. Am J Schaapherder AF, Johnson TA, Gettes LS, CKD: Insights from the RRI-CKD cohort Cardiol 77: 906–908, 1996 Janse MJ: Distribution of extracellular po- study. Clin J Am Soc Nephrol 5: 762–769, 21. Martinez-Vea A, Bardají A, Garcia C, Oliver tassium and its relation to electrophysio- 2010 JA: Severe hyperkalemia with minimal logic changes during acute myocardial 48. Gennari FJ, Segal AS: Hyperkalemia: An electrocardiographic manifestations: A re- ischemia in the isolated perfused porcine adaptive response in chronic renal in- port of seven cases. J Electrocardiol 32: 45– heart. Circulation 77: 1125–1138, 1988 sufficiency. Kidney Int 62: 1–9, 2002 49, 1999 37. Goyal A, Spertus JA, Gosch K, Venkitachalam 49. Salem MM, Rosa RM, Batlle DC: Extrarenal 22. O’Neil JP, Chung EK: Unusual electrocar- L, Jones PG, Van den Berghe G, Kosiborod M: potassium tolerance in chronic renal failure: diographic finding–bifascicular block due Serum potassium levels and mortality in acute Implications for the treatment of acute hy- to hyperkalemia. Am J Med 61: 537–540, myocardial infarction. JAMA 307: 157–164, perkalemia. AmJKidneyDis18: 421–440, 1976 2012 1991 23. Simon BC: Pseudomyocardial infarction 38. Grodzinsky A, Goyal A, Gosch K, 50. Sandle GI, Gaiger E, Tapster S, Goodship and hyperkalemia: A case report and sub- McCullough PA, Fonarow GC, Mebazaa A, TH: Evidence for large intestinal control of ject review. JEmergMed6: 511–515, 1988 Masoudi FA, Spertus JA, Palmer BF, potassium homoeostasis in uraemic pa- 24. Pothiawala SE: Hyperkalemia induced Kosiborod M: Prevalence and prognosis of tients undergoing long-term dialysis. Clin pseudo-myocardial infarction in septic hyperkalemia in patients with acute myo- Sci (Lond) 73: 247–252, 1987 shock. J Postgrad Med 60: 338–340, 2014 cardial Infarction. Am J Med 129: 858–865, 51. Grimm PR, Sansom SC: BK channels in the 25. Szerlip HM, Weiss J, Singer I: Profound hy- 2016 kidney. Curr Opin Nephrol Hypertens 16: perkalemia without electrocardiographic 39. Krogager ML, Eggers-Kaas L, Aasbjerg K, 430–436, 2007 manifestations. Am J Kidney Dis 7: 461– Mortensen RN, Køber L, Gislason G, Torp- 52. Foster ES, Jones WJ, Hayslett JP, Binder HJ: 465, 1986 Pedersen C, Søgaard P: Short-term mortal- Role of aldosterone and dietary potassium 26. Fordjour KN, Walton T, Doran JJ: Man- ity risk of serum potassium levels in acute in potassium adaptation in the distal colon of agement of hyperkalemia in hospitalized heart failure following myocardial in- the rat. Gastroenterology 88: 41–46, 1985 patients. Am J Med Sci 347: 93–100, 2014 farction. Eur Heart J Cardiovasc Pharmac- 53. Butterfield I, Warhurst G, Jones MN, Sandle 27. Dreyfuss D, Jondeau G, Couturier R, other 1: 245–251, 2015 GI: Characterization of apical potassium Rahmani J, Assayag P, Coste F: Tall T waves 40. Iseki K, Uehara H, Nishime K, Tokuyama K, channels induced in rat distal colon during during metabolic acidosis without hyper- Yoshihara K, Kinjo K, Shiohira Y, Fukiyama potassium adaptation. JPhysiol501: 537– kalemia: A prospective study. Crit Care K: Impact of the initial levels of laboratory 547, 1997 Med 17: 404–408, 1989 variables on survival in chronic dialysis pa- 54. Foley RN, Curtis BM, Randell EW, Parfrey 28. Huwez FU, Pringle SD, Macfarlane PW: tients. AmJKidneyDis28: 541–548, 1996 PS: Left ventricular hypertrophy in new he- Variable patterns of ST-T abnormalities in 41. Kovesdy CP, Regidor DL, Mehrotra R, Jing modialysis patients without symptomatic patients with left ventricular hypertrophy J, McAllister CJ, Greenland S, Kopple JD, cardiac disease. Clin J Am Soc Nephrol 5: and normal coronary arteries. Br Heart J 67: Kalantar-Zadeh K: Serum and dialysate po- 805–813, 2010 304–307, 1992 tassium concentrations and survival in he- 55. Bansal N, Keane M, Delafontaine P, Dries D, 29. Brady WJ, Chan TC: Electrocardiographic modialysis patients. Clin J Am Soc Nephrol Foster E, Gadegbeku CA, Go AS, Hamm LL, manifestations: Benign early repolarization. 2: 999–1007, 2007 Kusek JW, Ojo AO, Rahman M, Tao K, JEmergMed17: 473–478, 1999 42. Yusuf AA, Hu Y, Singh B, Menoyo JA, Wright JT, Xie D, Hsu CY; CRIC Study In- 30. Somers MP, Brady WJ, Perron AD, Mattu A: Wetmore JB: Serum potassium levels and vestigators: A longitudinal study of left The prominent T wave: Electrocardio- mortality in hemodialysis patients: A Retro- ventricular function and structure from CKD graphic . Am J Emerg spective Cohort study. Am J Nephrol 44: to ESRD: The CRIC study. Clin J Am Soc Med 20: 243–251, 2002 179–186, 2016 Nephrol 8: 355–362, 2013 31. An JN, Lee JP, Jeon HJ, Kim DH, Oh YK, 43. Torlén K, Kalantar-Zadeh K, Molnar MZ, 56. Bansal N, Fan D, Hsu CY, Ordonez JD, Go Kim YS, Lim CS: Severe hyperkalemia re- Vashistha T, Mehrotra R: Serum potassium AS: Incident atrial fibrillation and risk of quiring hospitalization: Predictors of mor- and cause-specific mortality in a large peri- death in adults with chronic kidney disease. tality. Crit Care 16: R225, 2012 toneal dialysis cohort. Clin J Am Soc J Am Heart Assoc 3: e001303, 2014 32. McMahon GM, Mendu ML, Gibbons FK, Nephrol 7: 1272–1284, 2012 57. Soliman EZ, Prineas RJ, Go AS, Xie D, Lash Christopher KB: Association between hy- 44. Genovesi S, Valsecchi MG, Rossi E, Pogliani JP, Rahman M, Ojo A, Teal VL, Jensvold perkalemia at critical care initiation and D, Acquistapace I, De Cristofaro V, Stella A, NG, Robinson NL, Dries DL, Bazzano L, mortality. Intensive Care Med 38: 1834– Vincenti A: Sudden death and associated Mohler ER, Wright JT, Feldman HI; Chronic 1842, 2012 factors in a historical cohort of chronic hae- Renal Insufficiency Cohort (CRIC) Study 33. Khanagavi J, Gupta T, Aronow WS, Shah T, modialysis patients. Nephrol Dial Trans- Group: Chronic kidney disease and preva- Garg J, Ahn C, Sule S, Peterson S: Hyper- plant 24: 2529–2536, 2009 lent atrial fibrillation: The Chronic Renal In- kalemia among hospitalized patients and 45. Brunelli SM, Du Mond C, Oestreicher N, sufficiency Cohort (CRIC). Am Heart J 159: association between duration of hyper- Rakov V, Spiegel DM: Serum potassium and 1102–1107, 2010 kalemia and outcomes. Arch Med Sci 10: short-term clinical outcomes among he- 58. Drawz PE, Babineau DC, Brecklin C, He J, 251–257, 2014 modialysis patients: Impact of the long Kallem RR, Soliman EZ, Xie D, Appleby D, 34. Friedensohn A, Faibel HE, Bairey O, interdialytic interval. Am J Kidney Dis 70: Anderson AH, Rahman M; CRIC Study In- Goldbourt U, Schlesinger Z: Malignant ar- 21–29, 2017 vestigators: Heart rate variability is a pre- rhythmias in relation to values of serum 46. Nakhoul GN, Huang H, Arrigain S, Jolly SE, dictor of mortality in chronic kidney disease: potassium in patients with acute myocardial Schold JD, Nally JV Jr., Navaneethan SD: A report from the CRIC Study. Am J Nephrol infarction. Int J Cardiol 32: 331–338, 1991 Serum potassium, end-stage renal disease 38: 517–528, 2013

J Am Soc Nephrol 28: 3155–3165, 2017 How Dangerous Is Hyperkalemia? 3163 BRIEF REVIEW www.jasn.org

59. Harnett JD, Foley RN, Kent GM, Barre PE, severe heart failure. N Engl J Med 341: 709– sodium zirconium cyclosilicate on potas- Murray D, Parfrey PS: Congestive heart 717, 1999 sium lowering for 28 days among outpa- failure in dialysis patients: Prevalence, in- 71. Rossignol P, Dobre D, McMurray JJ, tients with hyperkalemia: The HARMONIZE cidence, prognosis and risk factors. Kidney Swedberg K, Krum H, van Veldhuisen DJ, randomized clinical trial. JAMA 312: 2223– Int 47: 884–890, 1995 Shi H, Messig M, Vincent J, Girerd N, Bakris 2233, 2014 60. Rizk DV, Gutierrez O, Levitan EB, McClellan G, Pitt B, Zannad F: Incidence, determi- 81. Bushinsky DA, Williams GH, Pitt B, Weir MR, WM, Safford M, Soliman EZ, Warnock DG, nants, and prognostic significance of hy- Freeman MW, Garza D, Stasiv Y, Li E, Muntner P: Prevalence and prognosis of perkalemia and worsening renal function Berman L, Bakris GL: Patiromer induces unrecognized myocardial infarctions in in patients with heart failure receiving rapid and sustained potassium lowering in chronic kidney disease. Nephrol Dial the mineralocorticoid receptor antagonist patients with chronic kidney disease and Transplant 27: 3482–3488, 2012 eplerenone or placebo in addition to optimal hyperkalemia. Kidney Int 88: 1427–1433, 2015 61. Lorincz I, Mátyus J, Zilahi Z, Kun C, Karányi medical therapy: Results from the Eplerenone 82. DeFronzo RA: Hyperkalemia and hypo- Z, Kakuk G: QT dispersion in patients with in Mild Patients Hospitalization and Survival reninemic hypoaldosteronism. Kidney Int end-stage renal failure and during hemo- Study in Heart Failure (EMPHASIS-HF). Circ 17: 118–134, 1980 dialysis. JAmSocNephrol10: 1297–1302, Heart Fail 7: 51–58, 2014 83. Kamel KS, Ethier JH, Quaggin S, Levin A, 1999 72. Jain N, Kotla S, Little BB, Weideman RA, Albert S, Carlisle EJ, Halperin ML: Studies to 62. Sise ME, Courtwright AM, Channick RN: Brilakis ES, Reilly RF, Banerjee S: Predictors determine the basis for hyperkalemia in re- Pulmonary hypertension in patients with of hyperkalemia and death in patients with cipients of a renal transplant who are treat- chronic and end-stage kidney disease. cardiac and renal disease. Am J Cardiol ed with cyclosporine. J Am Soc Nephrol 2: Kidney Int 84: 682–692, 2013 109: 1510–1513, 2012 1279–1284, 1992 63. Kim SC, Chang HJ, Kim MG, Jo SK, Cho WY, 73. Tseng WC, Liu JS, Hung SC, Kuo KL, Chen 84. Dobbin SJH, Petrie JR, Lean MEJ, McKay Kim HK: Relationship between pulmonary YH, Tarng DC, Hsu CC: Effect of spi- GA: Fludrocortisone therapy for persistent hypertension, peripheral vascular calcifica- ronolactone on the risks of mortality and hos- hyperkalaemia. Diabet Med 34: 1005– tion, and major cardiovascular events in pitalization for heart failure in pre-dialysis 1008, 2017 dialysis patients. Kidney Res Clin Pract 34: advanced chronic kidney disease: A nation- 85. Sivakumar V, Sriramnaveen P, Krishna C, 28–34, 2015 wide population-based study. Int J Cardiol Manjusha Y, Reddy YS, Sridhar N, 64. Fenton F, Smally AJ, Laut J: Hyperkalemia 238: 72–78, 2017 Subramanian S: Role of fludrocortisone in and in a patient with kidney 74. Muto S, Sebata K, Watanabe H, Shoji F, the management of -induced hy- failure. Ann Emerg Med 28: 440–441, 1996 Yamamoto Y, Ohashi M, Yamada T, perkalemia in a renal transplant recipient. 65. Turgutalp K, Bardak S, Helvacı I, İs¸güzar G, Matsumoto H, Mukouyama T, Yonekura T, Saudi J Kidney Dis Transpl 25: 149–151, Payas E, Demir S, Kıykım A: Community- Namiki S, Kusano E: Effect of oral glucose 2014 acquired hyperkalemia in elderly patients: administration on serum potassium con- 86. Brown G: Fludrocortisone for heparin- Risk factors and clinical outcomes. Ren Fail centration in hemodialysis patients. Am J induced hyperkalemia. Can J Hosp Pharm 38: 1405–1412, 2016 Kidney Dis 46: 697–705, 2005 64: 463–464, 2011 66. Loutradis C, Tolika P, Skodra A, Avdelidou 75. Sica DA: Diuretic use in renal disease. Nat 87. Kaisar MO, Wiggins KJ, Sturtevant JM, A, Sarafidis PA: Prevalence of hyperkalemia Rev Nephrol 8: 100–109, 2011 Hawley CM, Campbell SB, Isbel NM, in diabetic and non-diabetic patients with 76. Lepage L, Dufour AC, Doiron J, Handfield Mudge DW, Bofinger A, Petrie JJ, Johnson chronic kidney disease: A Nested Case- K, Desforges K, Bell R, Vallée M, Savoie M, DW: A randomized controlled trial of flu- Control study. Am J Nephrol 42: 351–360, Perreault S, Laurin LP, Pichette V, Lafrance drocortisone for the treatment of hyper- 2015 JP: Randomized clinical trial of sodium kalemia in hemodialysis patients. Am J 67. Agarwal R, Sinha AD, Pappas MK, Abraham for the treatment of Kidney Dis 47: 809–814, 2006 TN, Tegegne GG: Hypertension in hemo- mild hyperkalemia in CKD. Clin J Am Soc 88.KimDM,ChungJH,YoonSH,KimHL:Ef- dialysis patients treated with atenolol or li- Nephrol 10: 2136–2142, 2015 fect of fludrocortisone acetate on reducing sinopril: A randomized controlled trial. 77. Weir MR, Bakris GL, Bushinsky DA, Mayo serum potassium levels in patients with end- Nephrol Dial Transplant 29: 672–681, 2014 MR, Garza D, Stasiv Y, Wittes J, Christ- stage renal disease undergoing haemodial- 68. Pitt B, Bakris G, Ruilope LM, DiCarlo L, SchmidtH,BermanL,PittB;OPAL-HK ysis. Nephrol Dial Transplant 22: 3273–3276, Mukherjee R; EPHESUS Investigators: Se- Investigators: Patiromer in patients with 2007 rum potassium and clinical outcomes in the kidney disease and hyperkalemia receiving 89. Jabbar AA, Wase A: Worsening wenck- Eplerenone Post-Acute Myocardial In- RAAS inhibitors. NEnglJMed372: 211– ebach after injection: farction Heart Failure Efficacy and Survival 221, 2015 Not uncommon but frequently missed di- Study (EPHESUS). Circulation 118: 1643– 78. Bakris GL, Pitt B, Weir MR, Freeman MW, agnosis. Am J Emerg Med 31(5): 893.e5– 1650, 2008 Mayo MR, Garza D, Stasiv Y, Zawadzki R, 893.e6, 2013 69. Pitt B, Remme W, Zannad F, Neaton J, Berman L, Bushinsky DA; AMETHYST-DN 90. Watanabe S, Shioda T, Kobayashi K, Ishizaki Martinez F, Roniker B, Bittman R, Hurley S, Investigators: Effect of patiromer on serum S, Ito F, Fujibayashi M, Tanaka M: Kleiman J, Gatlin M; Eplerenone Post-Acute potassium level in patients with hyper- cutis confined to the dermis after in- Myocardial Infarction Heart Failure Efficacy kalemia and diabetic kidney disease: The travenous administration of a calcium and Survival Study Investigators: Eplerenone, AMETHYST-DN randomized clinical trial. preparation: A case report and review of the a selective aldosterone blocker, in patients JAMA 314: 151–161, 2015 Japanese literature. Case Rep Dermatol 6: with left ventricular dysfunction after myo- 79. Packham DK, Rasmussen HS, Lavin PT, El- 85–90, 2014 cardial infarction. N Engl J Med 348: 1309– Shahawy MA, Roger SD, Block G, Qunibi W, 91. Moss J, Syrengelas A, Antaya R, Lazova R: 1321, 2003 Pergola P, Singh B: Sodium zirconium cy- : A of in- 70. Pitt B, Zannad F, Remme WJ, Cody R, closilicate in hyperkalemia. NEnglJMed travenous administration of calcium gluca- Castaigne A, Perez A, Palensky J, Wittes J; 372: 222–231, 2015 nate. J Cutan Pathol 33[Suppl 2]: 60–62, Randomized Aldactone Evaluation Study 80. Kosiborod M, Rasmussen HS, Lavin P, 2006 Investigators: The effect of spironolactone Qunibi WY, Spinowitz B, Packham D, Roger 92. Conte G, Dal Canton A, Imperatore P, De on morbidity and mortality in patients with SD, Yang A, Lerma E, Singh B: Effect of Nicola L, Gigliotti G, Pisanti N, Memoli B,

3164 Journal of the American Society of Nephrology J Am Soc Nephrol 28: 3155–3165, 2017 www.jasn.org BRIEF REVIEW

Fuiano G, Esposito C, Andreucci VE: Acute 97. Chou YH, Wang HY, Hsieh MS: Colonic hemodialysis. Arch Intern Med 141: 167– increase in plasma osmolality as a cause of necrosis in a young patient receiving oral 170, 1981 hyperkalemia in patients with renal failure. kayexalate in : Case report and lit- 102. Jadoul M, Thumma J, Fuller DS, Tentori F, Li Kidney Int 38: 301–307, 1990 erature review. Kaohsiung J Med Sci 27: Y, Morgenstern H, Mendelssohn D, Tomo 93. Chang KY, Lee IH, Kim GJ, Cho K, Park HS, 155–158, 2011 T, Ethier J, Port F, Robinson BM: Modifiable Kim HW: Plasma exchange successfully 98. McGowan CE, Saha S, Chu G, Resnick MB, practices associated with sudden death treats central pontine myelinolysis after Moss SF: Intestinal necrosis due to sodium among hemodialysis patients in the Dialysis acute from intravenous so- polystyrene sulfonate (Kayexalate) in sorbi- Outcomes and Practice Patterns study. Clin dium bicarbonate therapy. BMC Nephrol tol. South Med J 102: 493–497, 2009 J Am Soc Nephrol 7: 765–774, 2012 15: 56, 2014 99. Saginur M, Thiesen A, Lacson A, Bigam D, 103. Morrison G, Michelson EL, Brown S, 94. Cooper DJ, Walley KR, Wiggs BR, Russell Yap J: Small intestinal transplant mucosal Morganroth J: Mechanism and prevention JA: Bicarbonate does not improve hemo- necrosis associated with enteral sodium of cardiac arrhythmias in chronic hemodi- dynamics in critically ill patients who have polystyrene sulfonate administration. Am J alysis patients. Kidney Int 17: 811–819, . A prospective, controlled clini- Transplant 12: 3152–3154, 2012 1980 cal study. Ann Intern Med 112: 492–498, 1990 100. Bushinsky DA, Spiegel DM, Gross C, 104. Rossignol P, Legrand M, Kosiborod M, 95. Fox GN: Hypocalcemia complicating bi- Benton WW, Fogli J, Hill Gallant KM, Du Hollenberg SM, Peacock WF, Emmett M, carbonate therapy for salicylate poisoning. Mond C, Block GA, Weir MR, Pitt B: Effect of Epstein M, Kovesdy CP, Yilmaz MB, Stough West J Med 141: 108–109, 1984 patiromer on urinary excretion in WG, Gayat E, Pitt B, Zannad F, Mebazaa A: 96. Schiffl H: Sodium bicarbonate infusion for healthy adults. Clin J Am Soc Nephrol 11: Emergency management of severe hyper- prevention of : No evi- 1769–1776, 2016 kalemia: Guideline for best practice and dence for superior benefit, but risk for 101. Wiegand CF, Davin TD, Raij L, Kjellstrand opportunities for the future. Pharmacol Res harm? Int Urol Nephrol 47: 321–326, 2015 CM: Severe induced by 113[Pt A]: 585–591, 2016

J Am Soc Nephrol 28: 3155–3165, 2017 How Dangerous Is Hyperkalemia? 3165