BIOMARKERS OF OUTCOMES IN HEMODIALYSIS PATIENTS

Abstract

Chronic kidney disease is a complex condition, where the decrease in kidney function is accompanied by numerous metabolic changes affecting virtually all the organ systems of the human body. Many of the markers characteristic of the individually affected organ systems have been associated with adverse outcomes in CKD; it is believed that the high mortality seen in CKD is a result of several abnormalities conspiring to induce a heightened degree of morbidity and mortality. Not all the biomarkers may, however, be causally responsible for the adverse outcomes associated with them. We review various biomarkers of protein energy wasting, inflammation, oxidative stress, potassium disarrays, acid-base disorders, bone and mineral disorders, glycemic status, and anemia management. Although all of these biomarkers have shown associations with worsened outcomes in CKD, markers of protein and energy wasting, especially serum albumin, remain the strongest predictor of survival in CKD patients. We also review the putative pathophysiologic mechanisms behind these associations, and present possible therapeutic interventions that could result in remedies to improve poor clinical outcomes in CKD.

INTRODUCTION

Patients with ESRD have mortality rates that exceed those seen with most malignancies.¹ Observational studies have linked metabolic changes to increased morbidity and mortality in dialysis patients, but few clinical trials have been conducted to prove a causal link between these risk markers and adverse outcomes. We examined the various biochemical markers that have been associated with adverse outcomes in dialysis patients, and analyzed the potential interventions that may improve outcomes in this patient population.

Mineral and bone disorders

Disorders of mineral and bone metabolism occur frequently in CKD² and have been associated with increased mortality.^3;4 Observational studies in dialysis patients have linked hyperphosphatemia and the calcium-phosphorus product ^3;4 (Figure 1) and secondary hyperparathyroidism (SHPT)^3–7 with higher cardiovascular morbidity and mortality. The mechanism(s) responsible for these adverse outcomes could be phosphorus’ calcification-inducing effects,^8;9 or the deleterious effects of SHPT, which in itself is able to induce cardiovascular, metabolic, hematologic and immunologic changes.¹⁰ Vascular calcification was also found to be associated with higher serum calcium,^11–13 and higher serum calcium levels were associated with higher mortality ^3;4 in dialysis patients. Due to the lack of clinical trials the choice of the best therapeutic agent for treating the various aspects of CKD-MBD continues to be debated.^14;15

Figure 1.

Cardiovascular mortality associated with various levels of the calcium-phosphorus product in 58,058 maintenance hemodialysis patients. Based on data from Kalantar-Zadeh et al.⁴

Markers of anemia

Lower hemoglobin levels have been associated with adverse outcomes in MHD patients in multiple observational studies.^16–19 In clinical studies, moderate increase in hemoglobin concentration is associated with relief from symptoms of anemia and improved quality of life^18;20 and survival.²¹ Several recent randomized clinical trials have shown that targeting hemoglobin levels above 13 g/dL to “normalize” hemoglobin in CKD may be associated with poor clinical outcomes.^22–26 A recent retrospective study of 58,058 maintenance hemodialysis patients also demonstrated an inverse J-curve relationship between hemoglobin levels and adverse outcomes,¹⁹ in that hemoglobin levels between 11.5 and 13 g/dl were associated with the lowest death risk. Improved survival outcomes were observed in dialysis patients using erythropoesis stimulating agents (ESA) at any doses, whereas among those who received an ESA, a higher dose requirement was a surrogate of higher death risk, an effect that may be related to the association between ESA responsiveness index (averaged ESA dose divided by averaged hemoglobin within each 3 month) and malnutrition-inflammation complex (Figure 2).¹⁹ Finally, markers of iron status including serum ferritin and iron saturation ratio appear associated with survival in CKD patients,²⁷ although serum ferritin may also be an inflammatory marker.²⁸ The field of anemia of CKD is the only one where original observational studies of potential actionable risk factors in MHD were followed up by several large randomized controlled trials in an attempt to prove a causal link between the exposure (anemia) and the outcome (mortality of cardiovascular events).^22–26 Limitations inherent of many observational studies were thereby exposed, as higher hemoglobin targets in clinical trials have resulted in higher, not lower event rates. The reasons behind this are still debated, and include possible deleterious effects related to higher hemoglobin, but also to the higher doses of erythropoetin needed to achieve higher hemoglobin targets.^29–32 The debate has thus been centered not only on what the ideal hemoglobin level should be in MHD, but also on what the ideal treatment(s) for anemia is in these patients, with various novel ESA analogues and iron products making it potentially possible to explore alternative means of reaching the same goal.

Figure 2.

All-cause mortality associated with the ratio of recombinant human erythropoetin dose and blood hemoglobin level in 58,058 maintenance hemodialysis patients. Based on data from Regidor et al.¹⁹

Serum potassium

Both hypokalemia³³ and hyperkalemia^33;34 have been associated with higher mortality in patients on hemodialysis. The effect of hypokalemia on mortality appeared to be due to the poor nutritional status associated with this condition in a study of 81,013 patients receiving maintenance hemodialysis.³⁵ The importance of appropriate dialysate potassium concentrations was emphasized by the higher mortality seen in patients with hyperkalemia who were dialyzed with dialysates of higher, rather than lower potassium concentration.³⁵

Protein-energy wasting

Several markers of PEW have been associated with morbidity and mortality in patients on dialysis. Lower serum albumin,^36–39 prealbumin^40–46 and cholesterol^39;47–50 levels (Figure 3) have been associated with higher mortality in dialysis patients. Other biochemical markers that are directly or indirectly linked to PEW and outcomes include serum transferrin,^51;52 creatinine^53;54 and bicarbonate,^55;56 various hormones such as leptin,⁵⁷ visfatin,⁵⁸ adiponectin,⁵⁹ and thyroid hormones.⁶⁰ The mechanism of action responsible for the adverse outcomes associated with PEW markers remain unclear; it is likely that a combination of factors are responsible, rather than a single etiologic mechanism.⁶¹ Malnutrition causes impaired immune function leading to heightened susceptibility to infections and poor wound healing.⁶² Certain nutrients such as arginine and glutamine may enhance the immune response.^63–65 CKD patients may be susceptible to zinc,^66;67 vitamin B6 (pyridoxine), vitamin C and folic acid deficiencies,^68;69 which can induce alterations in host defense. Levocarnitine may protect against endotoxins and also suppress elaboration of tumor necrosis factor alpha (TNF-α) from monocytes.⁷⁰ Loss of adiposity in CKD could result in decreased sequestration of uremic toxins⁷¹ and lower production of anti-inflammatory cytokines and adiponectine.⁷² Sarcopenia seen in malnutrition results in reduced skeletal, respiratory and cardiac muscle function, and also in lower muscle oxidative metabolism with consequently decreased antioxidant defense.⁷³ Damage of muscle cells may cause the emergence of circulating actin that can consume gelsolin (which is in itself primarily produced by skeletal muscle), vitamin D binding protein and other circulating molecules with salutary and protective action.^74–76 The gastrointestinal tract is also adversely affected in malnutrition, with atrophy of the gut lining, decreased intestinal secretions and altered gut flora leading to further reduction in gut function and the ability to absorb nutrients.⁷⁷

Figure 3.

All-cause mortality associated with various levels of baseline serum cholesterol in 15,859 patients receiving maintenance hemodialysis. Based on data from Kilpatrick et al.⁵⁰

Some of the small clinical trials that have shown improvement in various nutritional markers have used oral or parenteral nutritional supplementation,^78–81 anabolic therapies with human growth hormone⁸² or testosterone supplementation,^83–85 or daily dialysis.^86;87 A recent randomized controlled trial of nutritional supplementation showed that an increase in prealbumin of >30 mg/L within 3 months predicted a 54% decrease in mortality, reduced hospitalizations and improved general well-being.⁸⁸ These results have to be interpreted conservatively, though, since the change in prealbumin was not the primary end point of the study.

Chronic inflammation

Chronic inflammation is one of many non-conventional risk factors that could explain the excess mortality in patients with CKD. Studies have reported a strong association between pro-inflammatory cytokine levels and adverse outcomes in CKD.^{52;54;89–91} A possible explanation for this is the involvement of inflammation in the process of atherosclerosis.⁹² The complexity of the inflammatory system has allowed the development of various antiinflammatory agents; unfortunately, most of these agents have not been studied in CKD yet, but the future application of some of them appears promising.⁹³

Oxidative stress

Markers of oxidative modification of lipids^94;95 and proteins^96;97 are present in increased amounts in patients with CKD. Low intake of dietary antioxidants has been associated with cardiovascular disease in the general population,⁹⁸ and various markers of oxidative stress were associated with higher mortality and cardiovascular disease in dialysis patients.^99–102 Small clinical trials in dialysis patients have shown lower cardiovascular event rates with the administration of vitamin E¹⁰³ and N-acetylcysteine.¹⁰⁴ Larger trials are needed to confirm these results and to determine if such interventions can decrease mortality.

Metabolic acidosis

In a study examining 56,385 maintenance hemodialysis patients a serum bicarbonate concentration <22 mEq/liter was associated with lower mortality in unadjusted analyses, but with higher mortality after extensive adjustments for markers of malnutrition and inflammation.⁵⁶ Lower bicarbonate was associated both with better nutritional status and worse outcomes in two other large observational studies of dialysis patients.^53;55 Treatment of metabolic acidosis by virtue of dialytic therapies is performed routinely, but no clinical trials have examined whether this interventions improves outcomes independent of the other effects of dialysis.

Glucose control

Several studies have examined outcomes associated with glycemic control in dialysis patients.^105–112 A recent large study of 24,875 dialysis patients¹⁰⁹ reported no association between HbA1c levels and mortality. The lack of survival association in this study could have been due to the short duration of follow-up (12 months) and inadequate controlling for malnutrition, inflammation and anemia. This was supported by another study that found a paradoxically lower unadjusted mortality associated with higher HbA1c levels, but adjustment for markers of malnutrition and inflammation resulted in higher HbA1c levels being associated with increased mortality.¹¹³ A single interventional study examined clinical outcomes in 83 dialysis patients undergoing intensive diabetes-related intervention compared to standard care.¹⁰⁷ Patients in the intensive intervention arm experienced improved quality of life and a reduction in the need for amputations and hospitalizations.¹⁰⁷ Establishment of specific blood glucose targets for dialysis patients is important, given that the complex glycemic dysregulation characteristic of the uremic state may make treatment guidelines derived from populations with normal kidney function inappropriate for such patients.¹¹⁴

CONCLUSIONS

Many biochemical markers are associated with adverse outcomes in CKD. Some of these represent factors that may be causally related to the adverse outcomes, and others could be merely markers of underlying complex pathophysiologic processes. Some therapeutic interventions in CKD have improved the level of various biomarkers, without definitive evidence for a beneficial effect on hard clinical end points such as mortality. Such effects need to be shown in future clinical trials.

Table.

Biomarkers	Putative mechanism of action of increased mortality	Therapeutic interventions
Nutritional markers: albumin, prealbumin, cholesterol, transferrin, creatinine, bicarbonate, leptin, visfatin, adiponectine, thyroid hormones, CRP, IL-6, TNF-α, serum amyloid A, lymphocyte count.	Complex, not well defined. May include effects mediated via lower muscle and fat mass, immune deficiency, insulin resistance and abnormal gastrointestinal function. Most biomarkers not causally involved in mechanism of action.	Oral or parenteral nutritional supplementation.a,c Anabolic therapies with human growth hormone or testosterone.c Daily dialysis.c Megestrol acetate.c
Markers of inflammation: CRP, IL-6, TNF-α, serum amyloid A, others.	Worsened atherosclerosis, endothelial dysfunction and oxidative stress.	Few studied explicitly for anti-inflammatory effect in CKD: Etanercept,c thiazolidinediones,c statins,c sevelamer hydrochloride.c
Markers of mineral and bone disorders: serum calcium, phosphorus, PTH, 25(OH) and 1,25(OH)2 vitamin D.	Vascular calcification. Various cardiovascular, metabolic, hematologic and immunologic abnormalities related to vitamin D metabolism and excess PTH.	Phosphate binders.b Treatment of secondary hyperparathyroidism with vitamin D-based therapiesc and calcium sensing receptor sensitizers.c
Serum potassium.	Cardiac arrhythmias.	Renal replacement therapy in dialysis patients.b Medical interventions affecting renal and extra-renal potassium excretion and cellular distribution.b
Serum bicarbonate.	Muscle wasting, decreased albumin synthesis, impaired cardiac function and glucose homeostasis, accumulation of beta-2 microglobulin, chronic inflammation and disturbances in growth hormone and thyroid function.	Bicarbonate replacement through oral and/or parenteral supplementation.c
Markers of glycemic control: blood glucose, glycosylated hemoglobin and albumin, fructosamine,	Complex mechanism of action involving micro- and macro-vascular, neurological and immunological mechanisms.	Multiple hypoglycemic therapies.3

^aMortality benefit in randomized controlled trial in CKD.

^bMortality benefit implied based on mechanism of action.

^cBenefit only proven on surrogate markers in CKD.