Regional citrate anticoagulation for continuous venovenous hemodiafiltration using calcium-containing dialysate1

Regional Citrate Anticoagulation for Continuous Venovenous Hemodiafiltration Using Calcium-Containing Dialysate Monika Gupta, MD, Nand K. Wadhwa, MD, and Rose Bukovsky, RN, BSN, CNN ● Background: Regional anticoagulation with trisodium citrate for continuous venovenous hemodiafiltration (CVVHDF) is an effective and safe alternative to heparin, especially in patients at high risk for bleeding. However, regional citrate anticoagulation is not used widely because current protocols are complex, labor intensive, and cumbersome. Existing protocols require the use of calcium-free dialysate with a continuous systemic calcium infusion to prevent hypocalcemia. We evaluated Anticoagulant Citrate Dextrose Formula A (ACD-A) solution for regional anticoagulation in CVVHDF in combination with a commercially available calcium-containing dialysis solution. Methods: Thirty-eight patients in the intensive care units underwent citrate-based CVVHDF using low-calcium peritoneal dialysis solution (calcium, 5.0 mg/dL [1.25 mmol/L]). ACD-A infusion rate was adjusted to maintain postfilter ionized calcium (iCaⴙⴙ) levels at 1.0 to 2.0 mg/dL (0.25 to 0.5 mmol/L). Calcium chloride (10%) solution was administered intravenously every 6 hours on an as-needed basis to maintain systemic serum iCaⴙⴙ levels at 3.5 to 4.0 mg/dL (0.88 to 1.0 mmol/L). Results: CVVHDF was performed for a total of 394 days using 149 hemofilters. Mean hemofilter life span was 63.5 ⴞ 27.1 hours. Seventy-five percent, 61%, and 49% of hemofilters were patent at 24, 48, and 72 hours, respectively. No patient experienced a change in clinical status caused by hypocalcemia and/or signs and symptoms of citrate toxicity. Four patients developed metabolic alkalosis requiring 0.1 N of hydrochloric acid infusion. Conclusion: Our simplified technique of regional citrate anticoagulation for CVVHDF using calcium-containing dialysate is not associated with increased hemofilter clotting and obviates the need for a continuous systemic calcium infusion and calcium-free dialysate. Am J Kidney Dis 43:67-73. © 2004 by the National Kidney Foundation, Inc. INDEX WORDS: Citrate anticoagulation; calcium-containing dialysate; Anticoagulant Citrate Dextrose Formula A (ACD-A); continuous venovenous hemodiafiltration (CVVHDF).

C

ONTINUOUS RENAL replacement therapy (CRRT) has been used increasingly for the management of renal failure in hemodynamically unstable and critically ill patients.1-6 The main disadvantage of CRRT is the need for anticoagulation to prevent clotting of the extracorporeal circuit, which further increases the risk for bleeding in this patient population. Currently, systemic heparin is used most commonly for anticoagulation in patients on CRRT.7 However, the major drawback of systemic heparin therapy is the risk for life-threatening bleeding episodes in the range of 25% to 30% in these patients.3,8,9 Saline flushes have been used to prevent clotting of the extracorporeal circuit, but their efficacy is unclear.3,9,10 Alternative methods for anticoagulation, namely, regional anticoagulation with heparin,11 low-molecular-weight heparin,12 prostacyclin,13-15 and the serine protease inhibitor nafamostat,16 are not used widely because of their limitations. Regional anticoagulation with citrate has been reported to be effective and safe, with lower bleeding risk.8,17-19 Citrate chelates ionized calcium (iCa⫹⫹) in the extracorporeal circuit. This causes anticoagulation of the extracorporeal circuit because iCa⫹⫹ is required for multiple steps in the clotting cascade. After citrate enters the

body, it is diluted by the total blood volume and rapidly metabolized to bicarbonate by the liver, skeletal muscles, and renal cortex. Regional anticoagulation with citrate increases the complexity of CRRT because of the need for a calcium-free dialysis solution and/or low-sodium replacement fluid to avoid metabolic complications and a systemic calcium infusion to prevent and/or treat hypocalcemia. We developed a protocol using calcium-containing peritoneal dialysis solution From the Department of Medicine, Division of Nephrology and Division of Nursing, State University of New York at Stony Brook, Stony Brook, NY. Received April 4, 2003; accepted in revised form September 3, 2003. Presented as a poster at the 8th International Conference on Continuous Renal Replacement Therapies, San Diego, CA, March 6-8, 2003, and 35th Annual Meeting and Scientific Exposition of the American Society of Nephrology, Philadelphia, PA, November 1-4, 2002. Published in abstract form in the J Am Soc Nephrol 13:605A, 2002 and Blood Purif 21:199, 2003. Address reprint requests to Nand K. Wadhwa, MD, Division of Nephrology, Department of Medicine, School of Medicine, HSC T-15 Rm-020, State University of NY at Stony Brook, Stony Brook NY 11794. E-mail: nwadhwa@ notes.cc.sunysb.edu © 2004 by the National Kidney Foundation, Inc. 0272-6386/04/4301-0008$30.00/0 doi:10.1053/j.ajkd.2003.09.014

American Journal of Kidney Diseases, Vol 43, No 1 (January), 2004: pp 67-73

67

68

GUPTA, WADHWA, AND BUKOVSKY Table 1.

Algorithm to Adjust the Rate of ACD-A Infusion

Posthemofilter iCa⫹⫹ (mg/dL)

⬎2.00 1.00-2.00 ⬍1.00

Change in ACD-A Infusion Rate

Increase rate by 10 mL/h No change in rate Decrease rate by 10 mL/h

NOTE. To convert iCa⫹⫹ from mg/dL to mmol/L, multiply by 0.25.

venous access. Postfilter iCa⫹⫹ samples were drawn from the blood sampling port distal to the hemofilter. Calcium chloride (10% solution; 13.6 mEq (6.8 mmol) of Ca⫹⫹/10 mL) was administered on an as-needed basis as an intermittent intravenous bolus every 6 hours to maintain systemic venous iCa⫹⫹ levels at 3.5 to 4.0 mg/dL (0.88 to 1.0 mmol/L; Table 2). Data are presented as mean ⫾ SD. t-Test and nonparametric Wilcoxon’s test were used when appropriate. P less than 0.05 is considered statistically significant.

RESULTS

in combination with Anticoagulant Citrate Dextrose Formula A (ACD-A) solution as the regional anticoagulant in continuous venovenous hemodiafiltration (CVVHDF), consequently eliminating the need for a customized dialysate and continuous systemic calcium infusion. We report our experience. METHODS We retrospectively reviewed the medical records of 38 patients with renal failure who had undergone citrate-based CVVHDF using calcium-containing dialysate in the intensive care units (ICUs) at our tertiary-care institution from January 2001 to September 2002. We evaluated the frequency of hemofilter clotting, hypocalcemia, hypernatremia, metabolic alkalosis, and citrate toxicity. Vascular access was established by insertion of a 12-Fr dual-lumen catheter (Arrow Int, Reading, PA) in either a femoral, internal jugular, or subclavian vein. Blood flow rate was set and maintained at 150 mL/min throughout the treatment. The Prisma M100 set with an AN69 hemofilter (Cobe, Lakewood, CO) was primed with 1 L of normal saline containing 5,000 IU of heparin. The Prisma M100 set was changed every 96 hours. Isotonic saline (sodium, 154 mEq/L [154 mmol/L]; chloride, 154 mEq/L [154 mmol/L]) was used as replacement fluid and infused proximal to the hemofilter at 500 to 1,000 mL/h. Magnesium sulfate and/or potassium chloride were added to the replacement fluid by the pharmacy at the discretion of the attending nephrologist. A low-calcium peritoneal dialysis solution (sodium, 132 mEq/L [132 mmol/ L]; calcium, 5.0 mg/dL [1.25 mmol/L]; chloride, 95 mEq/L [95 mmol/L]; lactate, 360 mg/dL [40 mmol/L]; magnesium, 0.5 mEq/L [0.25 mmol/L]; 1.5% dextrose, pH 5.2; osmolarity, 344 mOsm/L; Baxter Healthcare, Deerfield, IL) was used as dialysate. Dialysate was delivered at 500 to 1,000 mL/h. ACD-A (3% solution: trisodium citrate, 2.2%; and citric acid, 0.8%; 112.9 mmol of citrate/L; Baxter) was initiated at 150 mL/h through a Y connection at the junction of the double-lumen hemodialysis catheter and prefilter tubing. ACD-A rate was adjusted to maintain posthemofilter iCa⫹⫹ levels at 1.0 to 2.0 mg/dL (0.25 to 0.5 mmol/L; Table 1). Serum electrolytes, arterial blood gases, and complete blood cell count were measured at least twice a day and as needed. Systemic and posthemofilter iCa⫹⫹ were measured 1 hour after initiation of CVVHDF and then every 6 hours. Systemic iCa⫹⫹ samples were drawn from an existing intra-

Patient characteristics are listed in Table 3. Mean age was 55 ⫾ 16 years (range, 23 to 83 years). All patients were mechanically ventilated, hemodynamically unstable, and pressor dependent. Fluid overload was the most common indication for initiating CVVHDF. Thirty-three, 4, and 1 patients had a dual-lumen catheter in the femoral, internal jugular, and subclavian veins, respectively. CVVHDF was performed for a total of 394 days using 149 M 100 sets. Treatment variables during CVVHDF are listed in Table 4. The hemofilter lasted a mean of 63.5 ⫾ 27.1 hours. Twenty-six percent of hemofilters were patent at 96 hours. Twenty-six percent of hemofilters were discontinued before 96 hours for reasons other than clotting, and 25% were lost because of catheter malfunction. The censored mean survival of hemofilters was 74.2 ⫾ 24.3 hours (excluding filters lost because of catheter malfunction or reasons other than clotting). Seventy-five percent and 61% of hemofilters were patent at 24 and 48 hours, respectively (Fig 1). Eight patients had coagulopathy defined as a platelet count less than 75,000/␮L and/or international normalized ratio greater than 2.5. To exclude an artificial effect of coagulopathy on hemofilter survival, we performed a separate analysis of data from the remaining 30 patients. The censored hemofilter life span was 64.36 ⫾ 25.99 hours. This was not different in comparison to the group as a whole. There was no effect Table 2.

Schemata for Calcium Administration

Serum iCa⫹⫹ (mg/dL)

10% Calcium Chloride (mL)

3.5-4.0 3.0-3.4 ⬍3.0

0 10 20

NOTE. To convert iCa⫹⫹ from mg/dL to mmol/L, multiply by 0.25.

CITRATE ANTICOAGULATION FOR CVVHDF Table 3.

69 Patient Characteristics

Patient No.

Age (y)/Sex

Diagnosis

Indication for Citrate

Type of Renal Failure

Days on CVVHDF

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

47/M 55/M 47/M 52/M 74/M 36/F 66/F 54/M 32/M 32/M 59/M 50/M 50/M 57/M 76/M 56/M 64/M 23/M 52/M 78/M 30/M 83/M 30/F 50/M 23/M 72/F 80/M 44/M 72/F 70/F 50/M 51/M 74/M 82/F 55/M 65/M 53/F 47/M

Sepsis Sepsis CABG Sepsis CABG Sepsis Sepsis MVR Sepsis Sepsis Sepsis Craniotomy RAS GI bleed CABG CABG MVA MVA Sepsis Sepsis MVA Renal cell cancer Sepsis CABG Sepsis Lung cancer Sepsis Sepsis Lung cancer Colectomy Pancreatitis Colectomy CABG CABG Cholecystitis Endocarditis GI bleed Esophageal cancer

High risk for bleeding Thrombocytopenia Postoperative High risk for bleeding Postoperative High risk for bleeding Thrombocytopenia Postoperative DIC High risk for bleeding Thrombocytopenia Postoperative Postoperative Thrombocytopenia Postoperative Postoperative Postoperative High risk for bleeding High risk for bleeding Postoperative High risk for bleeding Postoperative High risk for bleeding Postoperative High risk for bleeding High risk for bleeding DIC Thrombocytopenia High risk for bleeding Postoperative Postoperative Postoperative Postoperative Postoperative Thrombocytopenia High risk for bleeding High risk for bleeding High risk for bleeding

ARF/CRF ARF/CRF ARF ARF ARF ARF ESRD ARF ARF ARF ARF ARF ARF ARF ARF ARF ARF ARF ARF ARF ARF ESRD ARF ESRD ARF ARF ARF ARF ARF ARF ARF ARF/CRF ARF/CRF ARF ARF ARF ARF/CRF ARF

26 14 4 5 15 5 4 3 5 11 21 5 10 4 3 2 2 3 4 6 40 12 18 4 4 3 2 6 8 20 25 5 6 13 46 9 16 5

Abbreviations: DIC, disseminated intravascular coagulopathy; MVR, mitral valve replacement; RAS, renal artery stenosis; ARF, acute renal failure; CRF, chronic renal failure; ESRD, end-stage renal disease; CABG, coronary artery bypass graft; MVA, motor vehicle accident; GI, gastrointestinal.

of underlying coagulopathy on hemofilter survival. Laboratory data at initiation and 48 hours after initiation of CVVHDF are listed in Table 5. Serum sodium levels were greater than 145 mEq/L (145 mmol/L) in 5 patients and greater than 150 mEq/L (150 mmol/L) in 1 patient. The lowest serum iCa⫹⫹ level observed was 2.8 mg/dL (0.7 mmol/L). No patient had a significant change in clinical status attributable to hypocal-

cemia. Serum lactate was measured in 18 patients. Mean serum lactate levels were 5.4 ⫾ 5.7 mEq/L (5.4 ⫾ 5.7 mmol/L) and 8.7 ⫾ 5.9 mEq/L (8.7 ⫾ 5.9 mmol/L) at initiation and 48 hours after initiation of CVVHDF, respectively. No patient developed signs of citrate toxicity, namely, an unexplained increase in anion gap and/or serum total calcium to serum iCa⫹⫹ ratio more than 2.5. Average serum bicarbonate level at 48 hours

70

GUPTA, WADHWA, AND BUKOVSKY Table 4.

Treatment Variables During CVVHDF

Variable

Mean ⫾ SD

ACD-A infusion rate (mL/h) Dialysate infusion rate (mL/h) Postfilter iCa⫹⫹ (mg/dL) Calcium administered per day (mEq/d [mmol/d])

159 ⫾ 10.78* 679.6 ⫾ 63.2 1.5 ⫾ 0.20 10.8 ⫾ 8.7 (5.4 ⫾ 4.35)

NOTE. To convert Ca⫹⫹ from mg/dL to mmol/L, multiply by 0.25. *Equals 0.97 ⫾ 1.22 mmol of citrate per hour and provides 1.99 ⫾ 0.14 mmol of citrate per/L of blood in the extracorporeal circuit.

was 26.83 ⫾ 6.54 mEq/L (26.83 ⫾ 6.54 mmol/ L). Four patients developed metabolic alkalosis requiring 0.1 N of hydrochloric acid (HCl) infusion. This was observed in our early attempts at citrate-based CVVHDF. There was no correlation between degree of alkalosis, hypernatremia, and citrate dose. The first patient developed metabolic alkalosis with a serum bicarbonate level of 36 mEq/L (36 mmol/L) after being on CVVHDF therapy for 10 days. This was related in part to acetate present in the total parental nutrition solution. Alkalosis was reversed by discontinuation of acetate and administration of 0.1 N of HCl, 100 mL/h, for 36 hours. The second patient developed a serum bicarbonate level of 34 mEq/L (34 mmol/L) after 6 days of CVVHDF therapy. This was related to nasal gastric aspiration in this patient with smallbowel obstruction. Alkalosis was corrected by infusion of HCl for 24 hours and the addition of histamine2 blockers (famotidine). The third and fourth patients developed serum bicarbonate lev-

els of 32 mEq/L (32 mmol/L) and 31 mEq/L (31 mmol/L), probably caused by increased hepatic production of bicarbonate from citrate and lactate metabolism. Both patients had been on CVVHDF therapy for 7 days. Mean ACD-A infusion rates were 170 and 180 mL/h for patients 3 and 4, respectively. Dialysate and replacement fluid infusion rates were 750 and 500 mL/h, respectively. Metabolic alkalosis was reversed by decreasing the dialysate rate to 500 mL/h and increasing the replacement fluid rate to 750 mL/h, along with the administration of 0.1 N of HCl for 12 hours. Although total exposure to the dialysate and replacement fluid remained constant, severe metabolic alkalosis was corrected by an increase in the convective loss of bicarbonate, along with reduced exposure of lactate from the dialysate. DISCUSSION

We report the use of ACD-A with calciumcontaining dialysate in CVVHDF, obviating the need for a continuous systemic calcium infusion. Sixty-one percent of hemofilters were patent at 48 hours. Mean life span of the hemofilter was 63.5 ⫾ 27.1 hours. Clinically significant hypocalcemia related to the use of citrate anticoagulation was not observed. Morita et al21 described the use of citrate for anticoagulation during intermittent hemodialysis in 1961. It was not until the late 1980s that regional citrate anticoagulation was first used for CRRT. Despite the simple basic principle behind regional citrate anticoagulation, current protocols are complex, and concerns about potential metabolic abnormalities exist. This makes hepa-

Fig 1. Percentage of functioning hemofilters over time.

CITRATE ANTICOAGULATION FOR CVVHDF Table 5.

71

Laboratory Data at Initiation and 48 Hours After CVVHDF

Variable

At Initiation

48 h After Treatment

P*

Sodium (mEq/L) Potassium (mEq/L) Chloride (mEq/L) Bicarbonate (mEq/L) Blood urea nitrogen (mg/dL) Creatinine (mg/dL) Magnesium (mg/dL) Total calcium (mg/dL) iCa⫹⫹ (mg/dL) Albumin (g/dL) PCO2 (mm Hg) Hemoglobin (g/dL) Platelet count (1,000/␮L) International normalized ratio Prothrombin time (s) Arterial pH Phosphorus (mg/dL)

136.68 ⫾ 8.27 4.93 ⫾ 1.2 104.72 ⫾ 7.43 19.63 ⫾ 5.15 64.85 ⫾ 47.02 3.63 ⫾ 2.24 2.24 ⫾ 0.54 7.65 ⫾ 0.55 3.97 ⫾ 0.48 2.11 ⫾ 0.67 41.69 ⫾ 14.01 9.86 ⫾ 1.07 140.48 ⫾ 97.09 2.10 ⫾ 0.82 14.84 ⫾ 2.44 7.34 ⫾ 0.11 5.20 ⫾ 2.74

137.95 ⫾ 16.62 4.32 ⫾ 0.51 97.79 ⫾ 8.52 26.83 ⫾ 6.54 51.28 ⫾ 28.63 2.83 ⫾ 1.53 1.94 ⫾ 0.30 7.58 ⫾ 0.54 3.47 ⫾ 0.38 2.14 ⫾ 0.70 44.59 ⫾ 15.14 9.46 ⫾ 1.43 129.34 ⫾ 81.39 2.41 ⫾ 1.23 15.42 ⫾ 3.31 7.33 ⫾ 0.13 4.65 ⫾ 1.60

NS ⬍0.01 ⬍0.01 ⬍0.01 NS NS ⬍0.05 NS ⬍0.001 NS NS NS NS NS NS NS NS

NOTE. Data expressed as mean ⫾ SD. To convert sodium, potassium, chloride, and bicarbonate in mEq/L to mmol/L, multiply by 1.0; albumin and hemoglobin in g/dL to g/L, multiply by 10; blood urea nitrogen in mg/dL to mmol/L, multiply by 0.357; creatinine in mg/dL to ␮mol/L, multiply by 88.4; magnesium in mg/dL to mmol/L, multiply by 0.4114; calcium in mg/dL to mmol/L, multiply by 0.2495; platelets to ⫻109/L, multiply by 1.0; and phosphorus in mg/dL to mmol/L, multiply by 0.3229. *P is significant at less than 0.05.

rin the favored method for anticoagulation in CRRT despite the lack of data showing differences in patient and hemofilter outcomes with the use of either heparin or citrate.8,20 In our study, average hemofilter life span was 63.5 ⫾ 27.1 hours (median, 72 hours), superior to the mean hemofilter survival of 51 hours reported by Chadha et al22 and median hemofilter survival of 29.5 ⫾ 17.9 hours, 39 hours, and 28.0 ⫾ 15.2 hours reported by Palsson et al,18 Tolwani et al,19 and Gabutti et al,23 respectively. Kutsogiannis et al17 reported a median survival of 82 hours using a similar system for CVVHDF, 3.9% trisodium citrate, and calcium-free dialysate. However, a modest number (24 filters) of filters were studied, and 15 of 24 filters were censored.17 Bunchman et al20 also reported a longer hemofilter survival of 71.3 ⫾ 7.2 hours using calcium-free dialysate and ACD-A for CVVHD. However, these investigators20 set the initial ACD-A infusion rate at 1.5 times the blood flow rate compared with 1:1 in the present study. The censored hemofilter life span was 64.36 ⫾ 25.99 hours in 30 patients without coagulopathy. There was no effect of underlying coagulopathy on hemofilter survival. There was a high rate of catheter malfunction causing premature loss of

the hemofilter, usually within the first few hours. Use of a femoral approach might be responsible for this degree of catheter malfunctioning. However, because of limited access options in this patient population, we continue to use a femoral approach. Currently, we are evaluating the effect of lower blood flow rates on catheter survival. Systemic iCa⫹⫹ levels ranged from 2.8 to 4.4 mg/dL (0.7 to 1.1 mmol/L), similar to those reported by Mehta et al8 (2.44 to 5.76 mg/dL [0.61 to 1.44 mmol/L]). No patient had a change in clinical status attributable to hypocalcemia. In addition, mean systemic iCa⫹⫹ level was 3.47 ⫾ 0.37 mg/dL (0.87 ⫾ 0.09 mmol/L), lower than the mean systemic iCa⫹⫹ level of 4.34 ⫾ 0.83 mg/dL (1.09 ⫾ 0.21 mmol/L) reported by Kutsogiannis et al.17 However, mean hourly calcium administered by Kutsogiannis et al17 was 3.7 ⫾ 1.4 mmol compared with 0.23 ⫾ 0.18 mmol in the present study. We believe the lower amount of calcium used in our study is largely caused by the use of calcium-containing dialysate. Nevertheless, a lower target for systemic iCa⫹⫹ and higher target for postfilter iCa⫹⫹ levels compared with other groups17,23 also might have contributed to the lesser use of calcium. Trisodium citrate can lead to various meta-

72

bolic disturbances, including hypernatremia8,21 and metabolic alkalosis.24 The incidence of hypernatremia in our study was similar to that observed by other investigators.17,20,23 Mean serum sodium level 48 hours after initiation of CVVHDF was 137.95 ⫾ 16.62 mEq/L (137.95 ⫾ 16.62 mmol/L; range, 130 to 154 mEq/L [130 to 154 mmol/L]). Only 1 patient had a serum sodium level greater than 150 mEq/L (150 mmol/L). In this patient, serum sodium levels normalized after the replacement fluid was switched from isotonic saline to 0.45% sodium chloride. The incidence of hypernatremia was lower than that reported by Mehta et al8 because there is lower sodium exposure with ACD-A compared with hypertonic citrate. The risk for alkalosis with ACD-A also is lower compared with hypertonic trisodium citrate because the former produces 33% less bicarbonate.25 In the present study, average serum bicarbonate level 48 hours after initiation of CVVHDF was 26.83 ⫾ 6.54 mEq/L (26.83 ⫾ 6.54 mmol/L). Four patients developed metabolic alkalosis requiring 0.1 N of HCl infusion. This was seen only in our initial patients undergoing citrate-based CVVHDF. Adjustment of dialysate and replacement fluid infusion rates and minimum use of acetate in total parental nutrition subsequently eliminated the development of metabolic alkalosis in these patients during CVVHDF using ACD-A solution as anticoagulant. No increase in arterial pH was seen despite an increase in serum bicarbonate levels. We calculated anion gap to evaluate for possible coexisting metabolic acidosis caused by the accumulation of unmeasured acids, namely, lactic acid and/or citric acid. Citrate can accumulate when the citrate infusion rate (ACD-A) exceeds hepatic metabolism and dialysate clearance of citrate. This is reflected by an unexplained anion gap acidosis, referred to as citrate gap. There was an increase in mean anion gap from 13.80 ⫾ 7.31 to 17.75 ⫾ 10.14 (P ⫽ not significant [NS]) at 48 hours after initiation of CVVHDF. Mean serum lactate levels increased from 5.4 ⫾ 5.7 mEq/L (5.4 ⫾ 5.7 mmol/L) to 8.7 ⫾ 5.9 mEq/L (8.7 ⫾ 5.9 mmol/L; P ⫽ NS) at 48 hours after initiation of CVVHDF. This increase in anion gap was explained mainly by the degree of lactic acidosis. There was no evidence of citrate gap. We believe

GUPTA, WADHWA, AND BUKOVSKY

lactic acidosis was caused primarily by tissue hypoperfusion in these critically ill patients, although the lactate in the dialysate also might have contributed. In addition, all patients were mechanically ventilated. Infusion of 0.9% sodium chloride as replacement fluid also might have counteracted the increase in pH to a degree because 0.9% sodium chloride is acidic (pH 5.4). In summary, results from this study show that ACD-A is an effective and safe regional anticoagulant for CVVHDF when used with calciumcontaining dialysate. Use of calcium-containing dialysate allows for the omission of a continuous calcium infusion and eliminates the need for a custom-made dialysis solution. ACKNOWLEDGMENT The authors thank nurse clinicians Mary J. Longobucco, RN, and Christine Teague, RN; dialysis staff; and ICU staff (Cardiovascular Intensive Care Unit, Medical ICU, Surgical ICU, and Coronary Care Unit) at University Hospital Medical Center, State University of New York at Stony Brook, NY, for their support in running a successful CVVHDF program.

REFERENCES 1. Kramer P, Wigger W, Rieger J, et al: Arteriovenous haemofiltration: A new and simple method for treatment of over-hydrated patients resistant to diuretics. Klin Wochenschr 55:1121-1122, 1977 2. Lauer A, Saccaggi A, Ronco C, et al: Continuous arteriovenous hemofiltration in the critically ill patient: Clinical use and operational characteristics. Ann Intern Med 99:455-460, 1983 3. Kaplan AA, Longnecker RE, Folkert VW: Continuous arteriovenous hemofiltration. A report of six months’ experience. Ann Intern Med 100:358-367, 1984 4. Schneider NS, Geronemus RP: Continuous arteriovenous hemodialysis. Kidney Int Suppl 33:S159-S162, 1988 5. Barzilay E, Weksler N, Kessler D, et al: The use of continuous arteriovenous hemodialysis in the management of patients with oliguria associated with multiple organ failure. Intensive Care Med 14:444-445, 1988 6. Bellomo R, Boyce N: Continuous venovenous hemodiafiltration compared with conventional dialysis in critically ill patients with acute renal failure. ASAIO J 39:M794M797, 1993 7. Abramson S, Niles JL: Anticoagulation in continuous renal replacement therapy. Curr Opin Nephrol Hypertens 8:701-707, 1999 8. Mehta RL, McDonald BR, Aguilar MM, et al: Regional citrate anticoagulation for continuous arteriovenous hemodialysis in critically ill patients. Kidney Int 38:976981, 1990 9. Ward DM, Mehta RL: Extracorporeal management of acute renal failure patients at high risk of bleeding. Kidney Int Suppl 43:S237-S244, 1993

CITRATE ANTICOAGULATION FOR CVVHDF

10. Paganini EP: Slow continuous hemofiltration and slow continuous ultrafiltration. ASAIO Trans 34:63-66, 1988 11. Kaplan AA, Petrillo R: Regional heparinization for continuous arterio-venous hemofiltration (CAVH). ASAIO Trans 33:312-315, 1987 12. Jeffrey RF, Khan AA, Douglas JT, et al: Anticoagulation with low molecular weight heparin (Fragmin) during continuous hemodialysis in the intensive care unit. Artif Organs 17:717-720, 1993 13. Langenecker SA, Felfernig M, Werba A, et al: Anticoagulation with prostacyclin and heparin during continuous venovenous hemofiltration. Crit Care Med 22:1774-1781, 1994 14. Canaud B, Mion C, Arujo A, et al: Prostacyclin (epoprostenol) as the sole antithrombotic agent in postdilutional hemofiltration. Nephron 48:206-212, 1988 15. Davenport A, Will EJ, Davison AM: Comparison of the use of standard heparin and prostacyclin anticoagulation in spontaneous and pump-driven extracorporeal circuits in patients with combined acute renal and hepatic failure. Nephron 66:431-437, 1994 16. Ohtake Y, Hirasawa H, Sugai T, et al: Nafamostat mesylate as anticoagulant in continuous hemofiltration and continuous hemodiafiltration. Contrib Nephrol 93:215-217, 1991 17. Kutsogiannis DJ, Mayers I, Chin WD, et al: Regional citrate anticoagulation in continuous venovenous hemodiafiltration. Am J Kidney Dis 35:802-811, 2000

73

18. Palsson R, Niles JL: Regional citrate anticoagulation in continuous venovenous hemofiltration in critically ill patients with a high risk of bleeding. Kidney Int 55:19911997, 1999 19. Tolwani AJ, Campbell RC, Schenk MB, et al: Simplified citrate anticoagulation for continuous renal replacement therapy. Kidney Int 60:370-374, 2001 20. Bunchman TE, Maxvold NJ, Barnett J, et al: Pediatric hemofiltration: Normocarb dialysate solution with citrate anticoagulation. Pediatr Nephrol 17:150-154, 2002 21. Morita Y, Johnson RW, Dorn RE, et al: Regional anticoagulation during hemodialysis using citrate. Am J Med Sci 242:32-42, 1961 22. Chadha V, Garg U, Warady BA, et al: Citrate clearance in children receiving continuous venovenous renal replacement therapy. Pediatr Nephrol 17:819-824, 2002 23. Gabutti L, Marone C, Colucci G, et al: Citrate anticoagulation in continuous venovenous hemodiafiltration: A metabolic challenge. Intensive Care Med 28:1419-1425, 2002 24. Mehta RL, McDonald BR, Ward DM: Regional citrate anticoagulation for continuous arteriovenous hemodialysis. An update after 12 months. Contrib Nephrol 93:210214, 1991 25. Flanigan MJ, Pillsbury L, Sadewasser G, et al: Regional hemodialysis anticoagulation: Hypertonic tri-sodium citrate or anticoagulant citrate dextrose-A. Am J Kidney Dis 27:519-524, 1996