Abstract
Background: Achieving “adequacy of dialysis” includes the maintenance of normal serum ionized calcium concentrations and is an important therapeutic goal in the treatment of acute renal failure (ARF). It is unknown whether this goal is best achieved with intermittent or continuous renal replacement therapy. Methods: We compared the effects of continuous veno–venous hemodiafiltration (CVVHDF) and intermittent hemodialysis (IHD) on serum ionized calcium concentrations using daily morning blood tests in 88 consecutive intensive care patients of which half were treated with IHD and half with CRRT. Results: Mean patient age was 54 ± 14 years for IHD and 60 ± 14 years for CVVHDF (NS). However, patients who received CVVHDF were significantly more critically ill (mean APACHE II scores: 24.4 ± 5.1 for IHD vs. 29.2 ± 5.7 for CVVHDF, p<0.003). Before treatment, the mean ionized calcium concentration was 1.177 ± 0.03 mmol/l for IHD and 1.172 ± 0.04 mmol/l for CVVHDF (NS), with abnormal values in 51.6% of IHD patients and in 68% of CVVHDF patients (NS). During treatment, hypocalcemia was significantly more common among CVVHDF patients (24.5% vs. 14.9%; p<0.011) while hypercalcemia was more frequent during IHD (36.1% vs. 25.6%; p<0.019). Conclusions: Abnormal serum ionized calcium concentrations are frequent in ARF patients before and during renal replacement. Once dialytic therapy is applied, CVVHDF is more likely to lower serum calcium concentrations, while IHD is more likely to induce hypercalcemia. Appreciation of these different biochemical effects may assist clinicians in adjusting dialytic therapy in selected patients.
INTRODUCTION
Acute renal failure (ARF) in the critically ill patient is still associated with a poor prognosis [[1]], [[2]] and, in the absence of randomized controlled trials, there is much controversy about which renal replacement therapy should be used [[3]], [[4]]. It has been suggested that IHD and continuous renal replacement techniques such as CVVHDF can be compared in terms of the “adequacy of dialysis” they deliver [[5]]. The importance of providing such “adequacy of dialysis” is supported by recent evidence suggesting that better uremic control may improve outcome [[6]]. On the other hand, the concept of dialytic adequacy does not solely relate to azotemic control. It should logically include other aspects of therapy such as hemodynamic stability, volume control, acid–base homeostasis, and the ability to deliver full nutritional support [[5]]. Last but not least, it should include the effects of renal replacement therapy (RRT) serum ionized calcium concentrations. We hypothesized that IHD and CVVHDF would affect ionized calcemia differently. Accordingly, we conducted such a comparative study in consecutive critically ill patients with ARF and now report our findings.
MATERIALS AND METHODS
This study is a retrospective controlled study. Forty-four critically ill patients with acute renal failure (ARF) treated with intermittent hemodialysis (IHD) in the intensive care unit (ICU) were studied. During this period, IHD was used as the only treatment for ARF in the ICU. These patients were compared to a further 44 similar consecutive patients treated with continuous veno–venous hemodiafiltration (CVVHDF) after a change of unit policy to the exclusive use of CVVHDF as renal support in the ICU. All patient records were reviewed to obtain demographic data, details of initial clinical presentation and biochemical data. For all patients, APACHE II [[7]] scores on admission and organ failure scores [[8]] at the time of initiation of renal replacement therapy were obtained. At the same time, daily morning biochemical measurements for the first fourteen days of treatment (on morning blood samples taken from patients for routine biochemical monitoring in the ICU) were retrieved from the computerized records of the Biochemistry Department. The focus of our data collection was the morning ionized calcium concentration. The reference ranges used by the laboratory for ionized calcium was 1.13 to 1.32 mmol/l. The Institutional Review Board waived the need for informed consent for this retrospective anonymous and confidential review of medical records.
Description of IHD Technique
IHD was applied according to the prescription of the attending nephrologist and was the only form of treatment available for ARF during the first part of the treatment period under study. The Nephrology team reviewed all patients at least once a day. This team included a consultant nephrologist, a Nephrology fellow, a resident and a dialysis nurse. Intermittent hemodialysis was prescribed according to the following principles:
Prevent hyperkalemia (K<6.0 mmol/l)
Prevent fluid overload
Avoid severe acidosis (pH<7.2)
Maintain plasma urea level at least<35 mmol/l and, if possible, <30 mmol/l most of the time
Avoid hemodynamic side effects of dialysis
In general, hemodialysis was performed 3 to 4 times per week for periods of 3 to 4 h per treatment session, according to the dialysis prescription. Vascular access was via double-lumen catheters or Scribner shunts. Blood flow was maintained between 200 and 300 ml/min as allowed by the catheter and dialysate flow rate was kept at 400 ml/min. Dialyzer surface area was variable and was adjusted to patient body size. Anticoagulation was achieved with heparin and occasionally omitted completely given the high bleeding risk of some patients. Most patients (n = 32) were dialyzed using bicarbonate-based dialysate. Acetate was used in close to 30% of patients (n = 12). The calcium concentration in the dialysate was kept between 1.5 and 1.75 mmol/l. All dialyzer membranes were made of cuprophane.
Description of CVVHDF Technique
CVVHDF was the sole CRRT modality throughout the second part of the study period due to a change of unit policy. CVVHDF was initiated and maintained by the intensivists and critical care nurses. Vascular access was obtained by insertion of a double lumen central venous catheter. Blood flow was maintained at 150 ml/min. Dialysate (Dianeal 1.5%, Baxter, Sydney, Australia) flow rate was maintained at 1 l/h. When clinically indicated, the dialysate flow rate was increased to 1.5 to 2 l/h (in the case of one patient) in order to increase azotemic control. Fluid replacement was administered at a rate determined by the spontaneous ultrafiltration rate (approximately 700 ml/h on average) and by frequent clinical assessment of the patients’ fluid status.
In most cases, the extracorporeal circuit required anticoagulation to prolong filter lifespan. Typically, this was achieved with low dose pre-filter heparin (500 IU/h) or with regional anticoagulation (full-dose pre-filter heparin with protamine reversal post-filter). Filter and patient aPTT were measured every 6–8 h and adjustments to the anticoagulation regime made accordingly. Anticoagulation was omitted altogether in patients at high risk of bleeding. Hemofilters were changed if clotting occurred or if the net ultrafiltration rate dropped below 150 ml/h for two consecutive hours.
Statistical Analysis
Summary descriptive statistics are presented as means ± SEM (standard error of the mean). When data were not normally distributed, comparisons were performed using the Mann-Whitney test, when normally distributed, comparisons were performed using Student's t-test. The Friedman's test (non-parametric analysis of variance) was used to determine if there was a significant change in the concentration of ionized calcium, which occurred during the course of renal replacement therapy in each treatment group. Post-hoc comparison of such changes, if detected, was performed using the Wilcoxon signed-rank test adjusted for multiple comparisons. Finally, Fisher's exact test was applied to determine if there was a significant difference in the incidence of abnormal (low or high or both) levels of ionized serum calcium in either group. The StatView™ (Abacus Concepts Inc., Berkeley, CA) statistical package was used for the above statistical analyses. A p<0.05 was considered statistically significant.
RESULTS
Eighty-eight patients were studied (44 treated with IHD and 44 treated with CVVHDF). The individual patient diagnoses are listed in Table 1. The two cohorts were similar in age and number of failing organs prior to commencement of renal replacement therapy. However, according to APACHE II scores, patients treated with CVVHDF were more severely ill than the IHD group (IHD: 24.4 ± 5.1 vs. CVVHDF: 29.2 ± 5.7; p<0.003) (Table 2). The two groups were otherwise similar in terms of treatment with inotropic drugs and mechanical ventilation. Both cohorts of patients also had similar incidences of bacteremia and fungemia.
Table 1. Major ICU Admission Diagnoses for the Two (IHD vs. CVVHDF) Treatment Groups
Table 2. Illness Severity in Treated Cohorts
The changes in morning mean ionized serum calcium concentration during therapy are presented in Figure 1. The serum ionized calcium concentration was frequently abnormal before treatment in both groups: 51.6% for IHD and 68% for CRRT (NS). Despite treatment, the ionized calcium concentration remained abnormal in 51% of observations during IHD and 50.2% of those during CVVHDF. However, abnormal values had a different pattern with the different therapies. During IHD, hypercalcemia was more common (36.1% of observations for IHD vs. 25.6% for CVVHDF; p<0.019), while during CVVHDF, hypocalcemia was more frequent (24.5% of observations for CVVHDF vs. 14.9% for IHD; p<0.011).
Published online:
07 July 2009Figure 1. Box-plot showing daily serum ionised calcium (mmol/l) levels from day 0 to day 13 during the course of treatment with both intermittent hemodialysis (IHD) (crossed boxes) and continuous veno–venous hemodiafiltration (CVVHDF) (grey boxes). The median value is displayed as the line within the box. The box represents 25th and 75th percentiles and the bars outside the box represent 10th and 90th percentiles. Circles outside the bars represent any outlying observations.
![]({"type":"image","src":"/na101/home/literatum/publisher/tandf/journals/content/irnf20/2002/irnf20.v024.i01/jdi-120002657/20150616/images/medium/irnf_a_11378997_uf0001_b.gif"})
Figure 1. Box-plot showing daily serum ionised calcium (mmol/l) levels from day 0 to day 13 during the course of treatment with both intermittent hemodialysis (IHD) (crossed boxes) and continuous veno–venous hemodiafiltration (CVVHDF) (grey boxes). The median value is displayed as the line within the box. The box represents 25th and 75th percentiles and the bars outside the box represent 10th and 90th percentiles. Circles outside the bars represent any outlying observations.
DISCUSSION
The dialytic management of ARF over the last decade has been characterized by the predominant use of either intermittent hemodialysis (IHD) or continuous renal replacement therapies (CRRT) [[3]], [[5]] and by controversy over their use [[9]]. The effect of these two types of RRT on mortality has not yet been compared in a randomized controlled study of suitable statistical power, therefore comparisons using surrogate markers have been used. One important surrogate marker of dialytic efficiency and safety is “adequacy of dialysis” [[5]], [[6]]. This concept encompasses the effect of RRT on azotemia, volume status, acid–base homeostasis, nutritional support and hemodynamic stability. Normalization of serum ionized calcium concentrations is another significant component of this conceptual framework. However, the differential effect of acute IHD or CRRT on serum ionized calcium levels has not yet been examined.
Biologically, calcium has important functions with concentrations regulated within a finite range. Abnormalities of serum ionized calcium levels may be associated with subtle but important adverse effects in patients with ARF. For example, calcium appears to have a central role in the pathogenesis of acute renal failure [[10]]. Experimentally a low-calcium environment or the use of calcium channel antagonists protects the tubules from renal injury, while a high calcium environment promotes tubular injury [[10]], [[11]]. Such effects of high serum calcium on tubular injury may be clinically significant and contribute to the aggravation or progression of renal injury [[12]]. Accordingly, the correction of hypercalcemia may be biologically important. Conversely, hypocalcemia is known to complicate critical illness in general and ARF. For example, hypocalcemic tetany induced by hyperphosphatemia has been reported in ARF [[13]] and close to 50% of critically ill patients may have hypocalcemia [[14]]. Correction of such hypocalcemia results in a significant and sustained improvement in blood pressure and left ventricular stroke work index [[14]].
Because of the potential adverse effects of disorders of calcemia an ideal dialytic technique should rapidly and reliably correct such disorders. Unfortunately, as demonstrated in this study, both IHD and CVVHDF fall short of such therapeutic goals. However, they do so in a different way. They both fail to correct disorders of calcemia during approximately half of the time with IHD being more commonly associated with hypercalcemia and CVVHDF with hypocalcemia. The full explanation for such differences is unclear but may derive, in part, from the use of a relatively high concentration of calcium in the dialysate during IHD in our patients and the well-known tendency to lose calcium via convective mechanisms during CVVHDF [[15]].
This study is retrospective with all the inherent shortcomings of retrospective studies. However, patients in both cohorts were similar in terms of clinical features, need for mechanical ventilation, inotropic drug therapy, number of failing organs, and incidence of bacteremia/fungemia. They also had similar disorders of serum ionized calcium before treatment. Therefore, this comparison of the differential effects of IHD and CVVHDF on serum ionized calcium concentrations is likely to reflect changes in larger populations. The morning concentration of serum ionized calcium was chosen as representative of its serum concentrations during therapy. Such values are objective, numerical and not subject to observer bias. We acknowledge that such concentrations are likely to fluctuate greatly. Nonetheless, there were hundreds of observations in the two groups, a number likely to ensure that the values used for this study were representative of the patterns and differences between the two forms of treatment. Indeed such differences emerged, were clear, had a high level of statistical significance, and are most unlikely to represent a type I error. Therefore, we believe that the findings of this investigation are broadly representative of the prevalent values during treatment.
It is possible that, if IHD had been applied for longer period of times or more frequently, the results of this comparison would have been different. However, second daily dialysis as reported in this study remains an accepted standard of practice for ARF in the year 2000, even in large world-renowned centers [[16]].
It is possible that different membranes would affect the serum ionized calcium concentration differently so that if biosynthetic membranes had been used, IHD would have achieved different results. On the other hand, no clinical data is available to sustain this view and the choice of membrane for ARF patients remains contentious [[16]]. Finally, acetate was used as buffer in a third of our patients. It is argued that bicarbonate buffering is preferable to acetate because of evidence of better hemodynamic stability in patients on chronic dialysis [[17]]. However, no randomized controlled trials have been performed in patients with ARF and the clinical significance of the possible hemodynamic benefits of bicarbonate remains unclear [[18]]. In relation to calcemia, differences between acetate and bicarbonate have never been documented.
In conclusion, we have described for the first time the incidence of disorders of ionized calcium in critically ill patients with ARF and their differential response to IHD and CVVHDF. Our findings demonstrate that such disorders are common and that they are only partly corrected by artificial renal support. If correction of these disorders is sought during renal replacement therapy, understanding how each therapy affects ionized calcium concentrations should help clinicians plan supplementation or treatment intensity.
ACKNOWLEDGMENTS
We wish to thank our medical, nursing and pathology staff at Monash Medical Centre, Clayton, Melbourne, Victoria, Australia, for their valuable contribution without which this paper would not have been possible. This study was supported by the Austin and Repatriation Medical Centre Intensive Care Research Fund.
| CVVHDF | IHD | |
|---|---|---|
| Post partum bleeding | 0 | 2 |
| Pneumonia | 6 | 4 |
| Pancreatitis | 2 | 4 |
| Ruptured abdominal aortic aneurysm | 4 | 4 |
| Hemolytic uremic syndrome | 2 | 1 |
| Empyema | 1 | 2 |
| Peritonitis | 6 | 8 |
| Necrotic gall bladder | 1 | 0 |
| Cellulitis | 1 | 0 |
| Vasculitis | 3 | 1 |
| Infarcted gut | 4 | 4 |
| Endocarditis | 1 | 2 |
| Meningitis | 1 | 2 |
| Myeloma | 1 | 0 |
| Cholesterol embolism | 1 | 0 |
| Heroin overdose/rhabdomyolysis | 1 | 0 |
| Urinary tract infection | 3 | 4 |
| Candidemia | 2 | 3 |
| Trauma | 1 | 3 |
| Massive operative bleeding | 2 | 0 |
| Ischio-rectal abscess | 1 | 0 |
| CVVHDF | IHD | p | |
|---|---|---|---|
| Number of patients | 44 | 44 | – |
| Mean age | 54±14 | 60±14 | NS |
| Mean APACHE II score | 29.2±5.7 | 24.4±5.1 | 0.003 |
| Number of failing organs pre-treatment | 3.5±1.5 | 4.0±1.4 | NS |
| Inotrope therapy | 35 | 36 | NS |
| Mechanical ventilation | 33 | 35 | NS |
| Bacteremia/fungemia | 16 | 13 | NS |
REFERENCES
- Spiegel D.M., Ullian M.E., Zerbe G.O., Berl T. Determinant of Survival and Recovery in Acute Renal Failure Patients Dialyzed in the Intensive Care Unit. Am. J. Nephrol. 1991; 11: 44–47 [Google Scholar]
- Brivet F., Kleinknecht D., Loirat P., Landais P. Acute Renal Failure in Intensive Care Units—Causes, Outcome, and Prognostic Factors: A prospective, Multicenter Study. Crit. Care Med. 1996; 24: 192–198 [Google Scholar]
- Lameire N., Van Biesen W., Vanholder R., Colardjin F. The Place of Intermittent Hemodialysis in the Treatment of Acute Renal Failure in the ICU Patient. Kidney Int. 1998; 53(S66)S110–S119 [Google Scholar]
- Bellomo R., Farmer M., Wright C., Parkin G., Boyce N. Treatment of Sepsis-Associated Severe Acute Renal Failure with Continuous Hemodiafiltration: Clinical Experience and Comparison with Conventional Dialysis. Blood Purif. 1998; 13: 246–254 [Google Scholar]
- Bellomo R., Ronco C. Acute Renal Failure in the Intensive Care Unit: Adequacy of Dialysis and the Case for Continuous Therapies. Nephrol. Dial. Transplant 1996; 11(3)424–428 [Google Scholar]
- Paganini E.P., Tapoliay M., Goormastic M. Establishing a Dialysis Therapy/Patient Outcome Link in Intensive Care Unit Acute Dialysis for Patients with Acute Renal Failure. Am. J. Kidney Dis. 1996; 28(Suppl 3)S81–S89 [Google Scholar]
- Knaus W.A., Drapner E.A., Wagner D.P., Zimmerman F.E. Apache II: A Severity of Disease Classification System. Crit. Care Med. 1985; 13: 818–829 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Fagon J.-Y., Chastre J., Novara A., Medioni P., Gilbert C. Characterization of Intensive Care Using a Model Based on the Presence or Absence of Organ Dysfunctions and/or Infections: The ODIN Model. Intensive Care Med. 1993; 19: 137–144 [Google Scholar]
- Kierdorf H.P., Sieberth H.G. Continuous Renal Replacement Therapies Versus Intermittent Hemodialysis in Acute Renal Failure: What Do We Know?. Am. J. Kidney Dis. 1996; 28(5S3)S90–S96 [Google Scholar]
- Edelstein C.L., Alkhunaizi A.A., Schrier R.W. The Role of Calcium in the Pathogenesis of Acute Renal Failure. Renal Failure 1997; 19: 199–207 [Taylor & Francis Online], [Google Scholar]
- Zager R.A. Pathogenetic Mechanisms in Nephrotoxic Acute Renal Failure. Seminars in Nephrol. 1997; 17: 3–14 [Google Scholar]
- Schneider M., Valentine S., Clarke G.M., Newman M.A., Peacock J. Acute Renal Failure in Cardiac Surgical Patients, Potentiated by Gentamicin and Calcium. Anesthesia and Intens. Care 1996; 24: 647–650 [Google Scholar]
- Zipser R.D., Bischel M.D., Abrams D.E. Hypocalcemic Tetany Due To Sodium Phosphate Ingestion in Acute Renal Failure. Nephron. 1975; 14: 378–381 [Google Scholar]
- Vincent J.L., Bredas P., Jankiwski S., Kahn R.J. Correction of Hypocalcemia in the Critically Ill: What Is The Hemodynamic Benefit?. Intensive Care Med. 1995; 21: 838–841 [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Sigler M.H., Teehan B.P. Solute Transport in Continuous Hemodialysis: A New Treatment For Acute Renal Failure. Kidney Int. 1987; 32(4)562–571 [Google Scholar]
- Albright R., Smelser J.M., Mccarthy J.T., Homburger H.A., Bergstrahl E.J., Larson T.S. Patient Survival and Renal Recovery in Acute Renal Failure: Randomized Comparison of Cellulose Acetate and Polysulfone Membrane Dialyzers. Mayo Clin. Proc. 2000; 75: 1141–1147 [Google Scholar]
- Iseki K., Onoyama K., Maeda T. Comparison of Hemodynamics Induced by Conventional Acetate Hemodialysis, Bicarbonate Hemodialysis and Ultrafiltration. Clin. Nephrol. 1980; 14: 294–301 [PubMed], [Web of Science ®], [Google Scholar]
- Henrich W.L., Woodard T.D., Meyer B.D., Chappell T.R., Rubin L.J. High Sodium Bicarbonate and Acetate Hemodialysis: Double Blind Crossover Comparison of Hemodynamic and Ventilatory Effects. Kidney Int. 1983; 24: 240–245 [Google Scholar]