Original ArticlesESRD and Chronic Dialysis

Serum Potassium and Cause-Specific Mortality in a Large Peritoneal Dialysis Cohort

Klara Torlén, Kamyar Kalantar-Zadeh, Miklos Z. Molnar, Tania Vashistha and Rajnish Mehrotra
CJASN August 2012, 7 (8) 1272-1284; DOI: https://doi.org/10.2215/CJN.00960112

Summary

Background and objectives Unlike hemodialysis (HD), peritoneal dialysis (PD) is a continuous therapy and does not induce myocardial stunning. Yet, the death risk in HD and PD patients is similar. This study tested the hypothesis that serum potassium abnormalities contribute more to the death risk in PD patients than in HD patients.

Design, setting, participants, & measurements Data from patients treated in DaVita facilities between July 1, 2001 and June 30, 2006 (n=10,468 PD patients; n=111,651 HD patients) were used to determine association of serum potassium with mortality.

Results PD patients were significantly more likely to have serum potassium <4 mEq/L, with an adjusted odds ratio of 3.30 (95% confidence interval [95% CI], 3.05, 3.56). There was a U-shaped relationship between time-averaged serum potassium and all-cause and cardiovascular mortality of PD patients, with adjusted hazards ratios of 1.51 for all-cause mortality for potassium <3.5 mEq/L (95% CI, 1.29, 1.76) and 1.52 for potassium ≥5.5 mEq/L (95% CI, 1.32, 1.75). The population-attributable risks for all-cause mortality for serum potassium <4.0 and ≥5.5 mEq/L were 3.6% and 1.9%, respectively, in PD patients, and 0.8% and 1.5%, respectively, in HD patients.

Conclusions Abnormalities in serum potassium contribute disproportionately to the high death risk in PD patients. This may, in part, account for the equivalent cardiac risk seen with the two therapies.

Introduction

The annual mortality of US dialysis patients exceeds 20%, and >40% of these deaths are secondary to cardiac causes. More than 60% of all cardiac deaths in dialysis patients are attributable to sudden cardiac death and/or arrhythmias (1). Many factors are likely operative in mediating the high cardiac risk seen in dialysis patients and include factors related to underlying diseases (hypertension, diabetes mellitus) and uremia (myocardial fibrosis) (2–5). In patients treated with hemodialysis (HD), the long interdialytic interval and the myocardial stunning induced by the HD procedure may possibly be additional contributing factors (6–8). Unlike HD, peritoneal dialysis (PD) is a continuous therapy and does not induce myocardial stunning (9). However, the incidence of cardiac deaths, including sudden cardiac death, is similar in PD patients to HD-treated patients (1).

Abnormalities in serum potassium are a risk factor that may, in part, account for the overall cardiac risk in PD patients to be of the same magnitude as that for HD. Thus, whereas hyperkalemia is commonly associated with advanced kidney disease and seen not infrequently in HD patients, clinical experience and small, single-center studies suggest that hypokalemia may be more common with PD. Furthermore, single-center studies from Taiwan (N=140) and Hong Kong (N=266) have shown an association between hypokalemia in PD patients and Enterobacteriaceae peritonitis and all-cause mortality, respectively (10,11). However, cultural differences influence dietary intakes of potassium (12). To our knowledge, distribution of serum potassium levels and its contribution to the high cardiac risk have not been investigated in a large and widely representative population of PD patients. Thus, we analyzed the data of all patients treated in facilities owned by DaVita Inc, a large provider of dialysis services in the United States, to test the hypothesis that hypokalemia is significantly more common in PD patients than in HD patients and contributes more to the high cardiac risk seen in PD patients than in HD patients.

Materials and Methods

Data Sources

This observational cohort study is based on data from maintenance dialysis patients treated in facilities owned by DaVita Inc between July 1, 2001 and June 30, 2006. During this period, 151,090 patients completed at least 90 days of treatment in facilities owned by DaVita Inc. The study cohort was stratified by dialysis modality, defined as the therapy being used at the time of entry into the cohort. Of the 151,090 participants, 7588 were excluded because of extremes in age (≤16 or ≥99 years). Of the 143,502 eligible participants (12,058 PD patients; 131,444 HD patients), data on serum potassium were missing for 21,383 individuals for the first 3-month period (1590 PD patients [13%]; 19,793 HD patients [15%]). This analysis is thus based on data from 122,119 participants (10,468 PD patients; 111,651 HD patients). The differences in characteristics of included and excluded patients, stratified by dialysis modality, are summarized in Supplemental Table 1.

The DaVita dataset was linked to that from the US Renal Data System (USRDS). Data from DaVita were used to determine a participant’s sex, age, presence of diabetes, body weight, and baseline height. The MEDEVID file from the USRDS contains data from the Centers for Medicare and Medicaid Services Form 2728, which is completed for all new ESRD patients in the United States and was used to determine the first day of dialysis, race/ethnicity, marital status, primary insurance, tobacco smoking, and seven comorbid conditions (atherosclerotic heart disease, including ischemic heart disease, myocardial infarction, and cardiac arrest; congestive heart failure; other cardiac diseases, including pericarditis and cardiac arrhythmia; cerebrovascular disease; peripheral vascular disease; chronic obstructive pulmonary disease; and cancer). Dialysis vintage was defined as the interval between the first day of dialysis and the first day of entering the cohort. Follow-up data were available through June 2007. The date and cause of death were obtained from USRDS.

Each patient had laboratory tests performed at least monthly, including serum potassium. All tests were performed in a single central laboratory located in Deland, Florida, within 24 hours of collection, and serum potassium was measured using an ion-sensitive electrode.

For each patient the average of all repeated measures done during the entry quarter was used in all statistical models (baseline data). The calendar quarter in which the patient completed 90 days of dialysis therapy was considered the baseline quarter. Averaged serum potassium was obtained for up to 20 calendar quarters to calculate the time-averaged serum potassium. Patients were divided into the following six groups to determine the relationship of serum potassium with outcome: <3.5, 3.5–<4.0, 4.0–<4.5, 4.5–<5.0, 5.0–<5.5, and ≥5.5 mEq/L. Additional analyses were done using serum potassium as a time-varying covariate.

The Institutional Review Board of the Los Angeles Biomedical Research Institute at Harbor-UCLA approved the study as exempt from informed consent.

Statistical Analyses

Data are summarized as means ± SD, medians (interquartile ranges), or proportions as appropriate. For the primary analysis, missing continuous data were imputed as the median and categorical variables as the prevalence in individuals with complete data. Data were complete for age, sex, diabetes, and race/ethnicity. Data for dialysis vintage, serum albumin, phosphorus, alkaline phosphatase, and hemoglobin were missing for <1%; serum creatinine, ferritin, total iron binding capacity, white blood cell count, and percentage of lymphocyte count were missing for 1%–2%; and comorbid conditions were missing for 5% of the cohort. Data for body mass index were missing for 7% of the cohort (PD, 42%; HD, 3%), primary insurance for 9%, serum parathyroid hormone for 12%, and marital status for 21%.

Logistic regression analysis was performed to determine the odds for PD patients to have baseline serum potassium below or above specific thresholds (<3.5, <4.0, ≥5.0, and ≥5.5 mEq/L), relative to HD patients. Survival analyses using Cox proportional hazards were utilized to determine the relationship of baseline and time-averaged serum potassium with all-cause, cardiovascular, and infection-related mortality in PD patients. Analyses with serum potassium as a time-dependent covariate were also done. Additional analyses were performed in the entire study cohort of PD and HD patients to compare the association of serum potassium with death risk between the two dialysis modalities, using time-averaged values and a time-dependent model. All outcomes were assigned to the initial dialysis modality. Patients were censored at time of transplantation. Secondary analyses were performed in which participants were censored at the time of transfer to another dialysis modality (PD to HD, or HD to PD, as appropriate). Data were analyzed using the following levels of adjustment: (1) unadjusted; (2) case-mix that included age, sex, race/ethnicity (Asians, Hispanics, blacks, whites, and others), seven comorbid conditions, history of tobacco smoking, categories of dialysis vintage (<6 months, 6 months to 2 years, 2–5 years, >5 years), primary insurance, and marital status; and (3) case-mix plus laboratory data that included all of the above covariates as well as body mass index, serum albumin, creatinine, ferritin, total iron binding capacity, calcium, phosphorus, parathyroid hormone, alkaline phosphatase, hemoglobin, white blood cell count, and percentage of lymphocyte count.

The population-attributable risk of death due to low and high time-averaged serum potassium in PD and HD patients was calculated according to the following equation: population-attributable risk = Pe × (RR − 1)/RR, where Pe is the prevalence of low or high serum potassium among all who died, and RR is the relative risk of death (13).

All analyses were carried out using STATA software (version 11.2; StataCorp LP, College Station, TX).

Results

Patient Characteristics

Compared with HD patients, PD patients were younger, were less likely to be black or widowed, were less likely to have diabetes mellitus or underlying cardiovascular comorbidities, and were more likely to be white and married (Table 1). Furthermore, PD patients had higher baseline serum creatinine, total iron binding capacity, and parathyroid hormone levels, and lower serum ferritin levels (Table 1).

View this table:
Table 1.

Baseline data of study cohort, stratified by modality

Distribution of Serum Potassium by Dialysis Modality

Figure 1 shows the distribution of baseline serum potassium by dialysis modality, and Table 2 summarizes the odds ratio of PD patients having values below and above clinically relevant thresholds. Serum potassium was <3.5 mEq/L in 10% of PD patients compared with 2% of HD patients, a difference in prevalence that persisted despite covariate adjustment (adjusted odds ratio, 4.17; 95% confidence interval [95% CI], 3.63, 4.79) (Table 2). Conversely, 20% of PD patients had serum potassium ≥5.0 mEq/L compared with 33% of HD patients (Figure 1). Thus, the adjusted odds for PD patients to have serum potassium ≥5.0 mEq/L were 44% lower (Table 2). Similar trends were observed for distribution of time-averaged serum potassium levels by dialysis modality.

Figure 1.
Figure 1.

Distribution of serum potassium levels in patients treated with PD (n=10,468) and HD (n=111,651). The data reflect the average of all measurements made during the first 3 months of entry into study cohort. PD, peritoneal dialysis; HD, hemodialysis.

View this table:
Table 2.

Odds ratios of PD patients having baseline serum potassium values below or above clinically relevant thresholds, using HD patients as the reference

Serum Potassium and Death Risk in PD Patients

Table 3 shows the characteristics of the 10,468 PD patients stratified by time-averaged serum potassium. With increasing serum potassium, patients were younger, were less likely to be female or black, and were less likely to have diabetes or cardiovascular comorbidities; however, they were more likely to have Medicare as their primary insurance. In addition, patients with higher serum potassium levels had longer dialysis vintage and higher serum albumin, creatinine, ferritin, phosphorus, parathyroid hormone, and alkaline phosphatase levels (Table 3).

View this table:
Table 3.

Baseline characteristics of patients treated with peritoneal dialysis at time of entry into cohort, stratified by time-averaged serum potassium

Follow-up data were available for 10,454 PD patients. The results showed that 4737 patients died over a median follow-up of 27 months. Patients with baseline serum potassium levels ≥5.0 mEq/L had a significantly higher adjusted risk for all-cause, cardiovascular, and infection-related mortality compared with those with levels between 4.0 and <4.5 mEq/L (Table 4). In contrast, there was no association between low baseline potassium and risk for either all-cause, cardiovascular, or infection-related mortality (Table 4).

View this table:
Table 4.

Association of baseline serum potassium and time-averaged serum potassium with all-cause, cardiovascular, and infection-related mortality in patients treated with PD at time of entry into the cohort (n=10,454)

In contrast, there was a U-shaped relationship between time-averaged serum potassium levels and all-cause and cardiovascular mortality (Table 4 and Figure 2). Thus, the magnitude of increase in the risk for all-cause mortality was similar in patients with time-averaged serum potassium levels <3.5 mEq/L (adjusted hazards ratio [HR], 1.51; 95% CI, 1.29, 1.76) to those with levels ≥5.5 mEq/L (adjusted HR, 1.52; 95% CI, 1.32, 1.75) (Table 4 and Figure 2). However, the increase in risk for infection-related mortality was seen only in individuals with time-averaged serum potassium <3.5 mEq/L (adjusted HR, 1.85; 95% CI, 1.32, 2.55) (Table 4). Similar trends were seen in each of the three major racial groups (Supplemental Table 2).

Figure 2.
Figure 2.

Association of time-averaged serum potassium concentration with mortality in patients undergoing PD (n=10,454). (A) All-cause mortality, (B) cardiovascular mortality, and (C) and infection-related mortality. Reference group: PD patients with time-averaged serum potassium levels between 4.0 and <4.5 mEq/L. Error bars represent 95% confidence intervals. PD, peritoneal dialysis.

When the participants were censored at the time of transfer to the other dialysis modality, the risk with low serum potassium levels was attenuated to a nonsignificant level, whereas the risk with higher levels persisted (Supplemental Table 3). The trend, with increased death risk with both low and high levels, was observed in models that used serum potassium as time-varying covariates (Figure 3 and Supplemental Table 4).

Figure 3.
Figure 3.

Association of serum potassium, use as a time-varying covariate, with mortality in patients undergoing peritoneal dialysis (n=10,454). Error bars represent 95% confidence intervals.

Comparison of Association of Serum Potassium with Death Risk in PD and HD Patients

Follow-up data were available for 10,454 PD patients and 111,434 HD patients. Whereas a higher risk for all-cause and cardiovascular mortality was seen in HD patients with serum potassium <4.0 or ≥5.0 mEq/L, higher risk was seen in those PD patients with levels <3.5 or ≥5.5 mEq/L (reference group: HD patients with serum potassium between 4.0 and <4.5 mEq/L) (Table 5 and Figure 4). In contrast, a higher risk for infection-related mortality was seen in HD patients with serum potassium <3.5 or ≥5.5 mEq/L, but in PD patients with time-averaged levels <4.5 or ≥5.5 mEq/L (Table 5).

View this table:
Table 5.

Association of time-averaged serum potassium and all-cause, cardiovascular, and infection-related mortality in patients treated with PD (n=10,454) and HD with dialysis modality defined at the time of entry into the cohort (n=111,434)

Figure 4.
Figure 4.

Association of time-averaged serum potassium concentration with outcomes in patients undergoing PD (n=10,454) and HD (n=111,434). (A) All-cause mortality, (B) cardiovascular mortality, and (C) infection-related mortality. Reference group: hemodialysis patients with serum potassium levels between 4.0 and <4.5 mEq/L. PD, peritoneal dialysis; HD, hemodialysis.

A similar trend, with increase in death risk with both low and high levels, was observed in analyses when patients were censored at the time of change in dialysis therapy (Supplemental Table 4) or in models that used serum potassium as time-varying covariates (Supplemental Table 5).

Population-Attributable Risk for Death to Low and High Serum Potassium Levels by Dialysis Modality

In patients treated with PD, the population-attributable risks of death associated with time-averaged serum potassium <4.0 and ≥5.5 mEq/L were 3.6% (95% CI, 1.7, 5.2) and 1.9% (95% CI, 1.3, 2.4), respectively. In patients treated with HD, the population-attributable risks of death associated with time-averaged serum potassium <4.0 and ≥5.5 mEq/L were 0.8% (95% CI, 0.6, 1.1) and 1.5% (95% CI, 1.3, 1.7), respectively.

Sensitivity Analyses

Two sets of sensitivity analyses were performed. In the first set, only individuals with complete data for all variables were included (PD, 4715; HD, 87,959). In the second set, analyses were repeated after excluding marital status and body mass index as covariates in the respective models. The same qualitative trends were observed for each analysis and are summarized in Supplemental Tables 6–8.

Discussion

To our knowledge, this is the first multicenter study to evaluate the distribution of serum potassium and its association with outcomes in PD patients. Our analyses of data from a large, contemporary cohort representative of the dialysis population in the United States indicates that over one-quarter of PD patients have time-averaged serum potassium <4.0 mEq/L, a level that is associated with a significant increase in risk for death in this population. Moreover, the population-attributable risk for all-cause mortality for serum potassium <4.0 mEq/L in PD patients was two-fold higher than with levels ≥5.5 mEq/L. Finally, the population-attributable risk for all-cause mortality for abnormalities in serum potassium levels in PD patients was two-fold higher than in those treated with HD, primarily due to substantially larger proportion of deaths associated with serum potassium <4.0 mEq/L.

Our study confirms the findings from previous small, single-center studies that low, rather than high, serum potassium levels are more common in PD patients. Thus, whereas 26% of patients had time-averaged serum potassium <4.0 mEq/L, 18% had levels ≥5.0 mEq/L. However, the prevalence of serum potassium <3.5 mEq/L (4% time-averaged values) in our cohort is lower than previously reported (7%–36%) from single-center studies (10,14–16). Our estimates likely have greater external validity because they were derived from a large multicenter cohort of PD patients. Nevertheless, other reasons for the varying estimates of prevalence may include differences in dietary intakes of potassium and PD prescription (volume and tonicity of dialysate). Furthermore, cohorts may differ in the use of oral potassium supplements; previous studies suggest that 10%–29% of PD patients require potassium supplements (15,17,18). The high prevalence of hypokalemia in ESRD patients treated with PD is likely directly attributable, at least in part, to the dialysis modality. None of the commercially available PD solutions have any added potassium. However, in a patient with ambient serum potassium concentration of 4 mEq/L and 13 L drain volume, the maximum potassium losses cannot exceed 52 mEq/d. The average daily intakes of potassium in the adult US population range from 54 to 86 mEq/d (12). Thus, dialysate potassium losses are unlikely to be sufficient to induce hypokalemia in most patients and likely require the presence of other factors such as inadequate dietary intakes and/or transcellular shifts induced by insulin release from absorption of glucose from peritoneal dialysate (19,20).

It is surprising that despite the well known association of both low and high serum potassium levels with all-cause and cardiovascular mortality in the general population, of hyperkalemia with kidney disease, and the high incidence of sudden cardiac death in dialysis patients, this is the first large-scale evaluation of the issue in patients treated with PD. Our study demonstrates a U-shaped relationship between time-averaged serum potassium and all-cause and cardiovascular mortality in PD-treated patients. Even though the association of low potassium levels with death risk was not seen with baseline values, time-averaged measures and time-dependent models provide more robust estimates of risk. A single-center study of Chinese patients from Hong Kong (n=266) previously demonstrated a higher risk for all-cause mortality in PD patients with serum potassium <3.5 mEq/L (10). However, unlike our study, Szeto et al. did not analyze association with cause-specific mortality or test the association of death risk with higher serum potassium levels. Thus, our findings build upon and extend those from the previous report. Given the large sample size used in our study and that the cohort was derived from centers distributed over most of the states in the United States, our findings also have considerable external validity (21,22).

To our knowledge, this is the first study to demonstrate an association between serum potassium and infection-related mortality in either PD- or HD-treated patients. Low time-averaged serum potassium was consistently associated with an increase in infection-related mortality in both PD and HD patients. In contrast, the higher risk for infection-related mortality with high serum potassium was evident in PD patients only when they were compared with HD patients with serum potassium between 4.0 and <4.5 mEq/L. Our findings are consistent with some previous observations about the association of low serum potassium levels with infectious risk in PD patients. In a recent study, Chuang et al. (11) found that the incidence of peritonitis was significantly higher among patients with hypokalemia; the risk was particularly high for infections with Enterobacteriaceae organisms. This is relevant because Gram-negative PD peritonitis is associated with significantly higher risk for either transfer to HD or mortality (23). Furthermore, a greater decline in serum potassium levels during the course of PD peritonitis has been reported to be associated with a higher likelihood of persistent infection (24). The biologic basis for the association of low serum potassium levels, however, either with risk for infection or infection-related mortality, is not known. It has been suggested that low serum potassium levels may alter intestinal motility, which, along with protein-energy wasting often associated with hypokalemia, may impair immunologic defenses. This, in turn, may lead to bacterial overgrowth and enhanced susceptibility to Enterobacteriaceae-associated peritonitis (10,11,25). However, these issues were not addressed in this study and remain speculative.

Even though the magnitude of death risk with low and high serum potassium levels is largely similar in both PD- and HD-treated patients, the importance of abnormal serum potassium levels differs by modality because of differences in prevalence. Thus, the population-attributable risk of deaths for levels of serum potassium <4.0 mEq/L was substantially larger in PD patients, because of a higher prevalence, compared with that in HD patients (26% versus 9%, respectively). Furthermore, in PD patients, the population-attributable risk for deaths with serum potassium <4.0 mEq/L was almost two-fold higher than seen with levels ≥5.5 mEq/L. These estimates indicate that abnormalities in serum potassium, particularly low levels, are a substantially more important contributor to death risk in PD patients than in HD patients.

Despite its strengths, our study has several important limitations. Observational cohort studies generate associations; it remains uncertain if these are causal. Despite adjustment for a large number of covariates, the risk for residual confounding remains. Furthermore, information about comorbidities was available only at the time of start of dialysis. Moreover, data on comorbidities were obtained from Centers for Medicare and Medicaid Services Form 2728 and there is concern that these data may underestimate the burden of coexisting diseases (26). Some patients had missing data on baseline serum potassium and they were excluded from analysis. Moreover, bias may have been introduced from missing data on some of the covariates used for adjustment. However, the proportion of patients with missing data on baseline serum potassium was small (15%). Moreover, we observed the same trends in death risk in our sensitivity analyses—both when the analyses were limited to those with complete data and those in which the statistical models did not include body mass index and marital status, the variables with the largest amount of missing data. Data on residual renal function, peritoneal transport type, PD modality, or prescription were unavailable; hence, we were unable to determine the association of these important variables with the distribution of serum potassium levels. Residual renal function facilitates the maintenance of potassium homeostasis and is an important determinant of death risk in dialysis patients (27). Our study cannot exclude the confounding influence of residual renal function on the association of higher serum potassium levels with death risk. Alternatively, hyperkalemia may be one of the mechanisms whereby loss of residual renal function increases death risk. Finally, data on medications, including potassium supplements and inhibitors of the renin-angiotensin-aldosterone system, were not available. This may have led us to underestimate the prevalence of low serum potassium in PD-treated patients; however, this is not important when testing the association with all-cause and cause-specific mortality.

In conclusion, our study demonstrates for the first time that abnormalities in serum potassium, particularly low levels, contribute substantially more to the death risk in PD-treated patients than in those treated with HD. Whether aggressive potassium supplementation would improve survival for these patients has yet not been tested. Until such studies are done, it may be prudent for physicians to carefully monitor and normalize potassium levels in PD-treated patients.

Disclosures

R.M. has received grant support and honoraria from Baxter Healthcare. The spouse of T.V. is an employee of DaVita Inc.

Acknowledgments

The study was supported by research grants from the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health (R21 DK 077341 to K.K.Z. and R.M.), a philanthropist grant from Mr. Harold Simmons (to K.K.Z.), and research grants from DaVita Clinical Research (to K.K.Z. and R.M.). M.Z.M. received grants from the National Developmental Agency (KTIA-OTKA-EU 7KP-HUMAN-MB08-A-81231) from the Research and Technological Innovation Fund, and is a recipient of the Hungarian Eötvös Scholarship (MÖB/77-2/2012).

Footnotes

  • Received January 26, 2012.
  • Accepted April 24, 2012.

References

  1. ↵
    1. National Institutes of Health, US Renal Data System
    : USRDS 2011 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States, Bethesda, MD, National Institutes of Health, 2011
  2. ↵
    1. Agarwal R
    : Blood pressure components and the risk for end-stage renal disease and death in chronic kidney disease. Clin J Am Soc Nephrol 4: 830–837, 2009pmid:19339424
    OpenUrlAbstract/FREE Full Text
    1. Gross ML,
    2. Ritz E
    : Hypertrophy and fibrosis in the cardiomyopathy of uremia—beyond coronary heart disease. Semin Dial 21: 308–318, 2008pmid:18627569
    OpenUrlCrossRefPubMed
    1. Duong U,
    2. Mehrotra R,
    3. Molnar MZ,
    4. Noori N,
    5. Kovesdy CP,
    6. Nissenson AR,
    7. Kalantar-Zadeh K
    : Glycemic control and survival in peritoneal dialysis patients with diabetes mellitus. Clin J Am Soc Nephrol 6: 1041–1048, 2011pmid:21511838
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Kalantar-Zadeh K,
    2. Kopple JD,
    3. Regidor DL,
    4. Jing J,
    5. Shinaberger CS,
    6. Aronovitz J,
    7. McAllister CJ,
    8. Whellan D,
    9. Sharma K
    : A1C and survival in maintenance hemodialysis patients. Diabetes Care 30: 1049–1055, 2007pmid:17337501
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Foley RN,
    2. Gilbertson DT,
    3. Murray T,
    4. Collins AJ
    : Long interdialytic interval and mortality among patients receiving hemodialysis. N Engl J Med 365: 1099–1107, 2011pmid:21992122
    OpenUrlCrossRefPubMed
    1. Breidthardt T,
    2. McIntyre CW
    : Dialysis-induced myocardial stunning: The other side of the cardiorenal syndrome. Rev Cardiovasc Med 12: 13–20, 2011pmid:21546884
    OpenUrlPubMed
  8. ↵
    1. Burton JO,
    2. Jefferies HJ,
    3. Selby NM,
    4. McIntyre CW
    : Hemodialysis-induced cardiac injury: Determinants and associated outcomes. Clin J Am Soc Nephrol 4: 914–920, 2009pmid:19357245
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Selby NM,
    2. McIntyre CW
    : Peritoneal dialysis is not associated with myocardial stunning. Perit Dial Int 31: 27–33, 2011pmid:20522673
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Szeto CC,
    2. Chow KM,
    3. Kwan BC,
    4. Leung CB,
    5. Chung KY,
    6. Law MC,
    7. Li PK
    : Hypokalemia in Chinese peritoneal dialysis patients: Prevalence and prognostic implication. Am J Kidney Dis 46: 128–135, 2005pmid:15983966
    OpenUrlCrossRefPubMed
  11. ↵
    1. Chuang YW,
    2. Shu KH,
    3. Yu TM,
    4. Cheng CH,
    5. Chen CH
    : Hypokalaemia: An independent risk factor of Enterobacteriaceae peritonitis in CAPD patients. Nephrol Dial Transplant 24: 1603–1608, 2009pmid:19103738
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Kant AK,
    2. Graubard BI,
    3. Kumanyika SK
    : Trends in black-white differentials in dietary intakes of U.S. adults, 1971-2002. Am J Prev Med 32: 264–272, 2007pmid:17383557
    OpenUrlCrossRefPubMed
  13. ↵
    Greenland S: Applications of stratified analysis methods. In: Modern Epidemiology, 2nd ed., edited by Rothman K, Greenland S, Philadelphia, Lippincott-Raven, 1998
  14. ↵
    1. Spital A,
    2. Sterns RH
    : Potassium supplementation via the dialysate in continuous ambulatory peritoneal dialysis. Am J Kidney Dis 6: 173–176, 1985pmid:3898826
    OpenUrlPubMed
  15. ↵
    1. Tziviskou E,
    2. Musso C,
    3. Bellizzi V,
    4. Khandelwal M,
    5. Wang T,
    6. Savaj S,
    7. Oreopoulos DG
    : Prevalence and pathogenesis of hypokalemia in patients on chronic peritoneal dialysis: One center’s experience and review of the literature. Int Urol Nephrol 35: 429–434, 2003pmid:15160552
    OpenUrlCrossRefPubMed
  16. ↵
    Jung JY, Chang JH, Lee HH, Chung W, Kim S: De novo hypokalemia in incident peritoneal dialysis patients: A 1-year observational study. Electrolyte Blood Press 7: 73–78, 2009
  17. ↵
    1. Oreopoulos DG,
    2. Khanna R,
    3. Williams P,
    4. Vas SI
    : Continuous ambulatory peritoneal dialysis - 1981. Nephron 30: 293–303, 1982pmid:7110460
    OpenUrlCrossRefPubMed
  18. ↵
    1. Khan AN,
    2. Bernardini J,
    3. Johnston JR,
    4. Piraino B
    : Hypokalemia in peritoneal dialysis patients. Perit Dial Int 16: 652, 1996pmid:8981546
    OpenUrlFREE Full Text
  19. ↵
    1. Clausen T,
    2. Everts ME
    : Regulation of the Na,K-pump in skeletal muscle. Kidney Int 35: 1–13, 1989pmid:2540370
    OpenUrlCrossRefPubMed
  20. ↵
    1. Hundal HS,
    2. Marette A,
    3. Mitsumoto Y,
    4. Ramlal T,
    5. Blostein R,
    6. Klip A
    : Insulin induces translocation of the alpha 2 and beta 1 subunits of the Na+/K(+)-ATPase from intracellular compartments to the plasma membrane in mammalian skeletal muscle. J Biol Chem 267: 5040–5043, 1992pmid:1312081
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Lowrie EG,
    2. Lew NL
    : Death risk in hemodialysis patients: The predictive value of commonly measured variables and an evaluation of death rate differences between facilities. Am J Kidney Dis 15: 458–482, 1990pmid:2333868
    OpenUrlPubMed
  22. ↵
    1. Iseki K,
    2. Uehara H,
    3. Nishime K,
    4. Tokuyama K,
    5. Yoshihara K,
    6. Kinjo K,
    7. Shiohira Y,
    8. Fukiyama K
    : Impact of the initial levels of laboratory variables on survival in chronic dialysis patients. Am J Kidney Dis 28: 541–548, 1996pmid:8840944
    OpenUrlPubMed
  23. ↵
    1. Szeto CC,
    2. Chow VC,
    3. Chow KM,
    4. Lai RW,
    5. Chung KY,
    6. Leung CB,
    7. Kwan BC,
    8. Li PK
    : Enterobacteriaceae peritonitis complicating peritoneal dialysis: A review of 210 consecutive cases. Kidney Int 69: 1245–1252, 2006pmid:16467787
    OpenUrlCrossRefPubMed
  24. ↵
    1. Murata GH,
    2. Fox L,
    3. Tzamaloukas AH
    : Predicting the course of peritonitis in patients receiving continuous ambulatory peritoneal dialysis. Arch Intern Med 153: 2317–2321, 1993pmid:8215734
    OpenUrlCrossRefPubMed
  25. ↵
    1. Kovesdy CP,
    2. Regidor DL,
    3. Mehrotra R,
    4. Jing J,
    5. McAllister CJ,
    6. Greenland S,
    7. Kopple JD,
    8. Kalantar-Zadeh K
    : Serum and dialysate potassium concentrations and survival in hemodialysis patients. Clin J Am Soc Nephrol 2: 999–1007, 2007pmid:17702709
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Longenecker JC,
    2. Coresh J,
    3. Klag MJ,
    4. Levey AS,
    5. Martin AA,
    6. Fink NE,
    7. Powe NR
    : Validation of comorbid conditions on the end-stage renal disease medical evidence report: The CHOICE study. Choices for Healthy Outcomes in Caring for ESRD. J Am Soc Nephrol 11: 520–529, 2000pmid:10703676
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Bargman JM,
    2. Thorpe KE,
    3. Churchill DN,
    4. CANUSA Peritoneal Dialysis Study Group
    : Relative contribution of residual renal function and peritoneal clearance to adequacy of dialysis: A reanalysis of the CANUSA study. J Am Soc Nephrol 12: 2158–2162, 2001pmid:11562415
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top

In this issue

View Selected Citations (0)
Print
Download PDF
Article Alerts
Email Article
Citation Tools
Request Permissions
Serum Potassium and Cause-Specific Mortality in a Large Peritoneal Dialysis Cohort
Klara Torlén, Kamyar Kalantar-Zadeh, Miklos Z. Molnar, Tania Vashistha, Rajnish Mehrotra
CJASN Aug 2012, 7 (8) 1272-1284; DOI: 10.2215/CJN.00960112
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

More in this TOC Section

Cited By...

Similar Articles

Related Articles