Paracentesis with Dextran 70 vs. paracentesis with albumin in cirrhosis with tense ascites

310

@ 1992 Elsevbr Science

Journal of Hepatology, 1992;14: 310-316 PublishersB.V. AHrights reserved. 0168-8278/92/$05.00

HEPAT 01091

Paracentesis with Dextran 70 vs. aracentesis wit with tense ascites

albumin in cirr

Results of a randomized study Eduardo Fassio’, Ruben Terg2, Graciela Landeira’, Raquel Abecasis2, Marta Salemnel, Ana Podesta2, Patricia Rodriguez’, Diana Levi2 and David Kravetz’ 'Hospital profAlejandro posadas, ‘Hospital National de Gastroenterologia Dr. Bonorino Udaondo, Buenos Aires, Argentina

(Received 3 February 1991)

Forty-one patients with cirrhosis and tense ascites were randomized to receive daily paracentesis of 5 liters associated with Dextran 70 as volume expander (6 g for each 1000 ml of ascites removed) (group I = 20 patients) or paracentesis with albumin (6 g for each 1000 ml of ascites) (group II = 21 patients). The basal clinical features, laboratory data, and plasmarenin activity were similar in both groups. The volume of ascites removed was 12.9 + 4.4 and 10.9 + 3.7 liters in group I and II, respectively (n.s.). No significant changes were observed in liver and renal function tests, KP’IT, platelet count, factor VIII, serum electrolytes or plasma renin activity 24 and 96 h after the last paracentesis in both groups, except for a decrease in bilirubin in group I and a transient increase of serum albumin in group II. Four patients developed complications in each group, mainly hyponatremia, while one patient in each group developed renal impairment. One patient from group I died with hepatic encephalopathy. Moreover, the probability of survival and readmission to the hospital because of tense ascites were similar in both groups of patients during the follow-up. The treatment cost with Dextran 70 was 15.50 dollars vs. 364.30 dollars with albumin for each patient treated. These results indicate that repeated large volume paracentesis associated with Dextran 70 is as effective and safe as paracentesis associated with albumin in cirrhotic patients with tense ascites. However, due to its reduced cost, paracentesis with Dextran 70 may be considered the treatment of choice in cirrhotic patients with tense ascites without liver cancer and renal failure.

It has recently been shown that paracentesis associated with intravenous albumin infusion is an effective and safe therapy for ascites in cirrhotic patients. It is associated with a similar or lower incidence of complications than diuretic treatment and considerably shortens the duration of hospital stay (l-3). Gines et al. compared therapeutic paracentesis with and without intravenous albumin infusion and showed that the administration of albumin avoided renal and electrolyte complications and the activation of endogenous vasoactive systems (4). Considering the high cost of albumin, we decided to investigate the effectiveness of therapeutic paracentesis associated with a plasma volume expander of low cost

such as Dextran 70. In this controlled and randomized study we compared the effectiveness, safety and cost of repeated paracentesis with Dextran 70 vs. repeated paracentesis with albumin in the treatment of cirrhotic patients with tense ascites.

Patients and Methods We studied 41 cirrhotic patients with tense ascites hospitalized in the Hospital National Profesor t?slejandro Posadas and in the Hospitai National de Gastroenterologia Carlos B. I_Jdaondo between April 1985 and March

Correspondence:Dr. Ruben Terg, Hospital National de Gastroe!lteroIogia Argentina.

Dr. Bonorino

Udaondo,

Caseros 2061, (1264) Bueaos Aires,

PARACENTESIS

WITH DEXTRAN

IN CIRRHOTIC

PATIENTS

1990. The diagnosis of cirrhosis was based on liver biopsy or on clinical, laboratory and ultrasonography findings, The exclusion criteria were: (i) primary or secondary liter cancer; (ii) gastrointestinal hemorrhage; (iii) serum urea > 60 mg/dl or serum creatinine > 1.5 mg/dl; (iv) infection; (v) encephalopathy; (vi) prothrombin concentration < 40%; (vii) platelet count < 40.000/mm3; (viii) serum bilirubin > 10 mg/dl; and (ix) urinary sodium concentration > 20 mEq/l. The etiology of cirrhosis was alcoholic in 34 patients, hepatitis B virus in two, NANB hepatitis in two and cryptogenic in three. Thirty-one patients were men. The study was approved b:; the Investigation Committee of each hospital and all patients gave informed consent. After admission, the patients received a diet containing 50 mEq of sodium per day. Hyponatremic patients had water restriction, and diuretic treatment was withdrawn in all patients for at least 5 days before the first paracentesis. Patients were then randomly allocated to group I (paracentesis with Dextran) or group II (paracentesis with albumin). There were 20 patients in group I and 21 in group II. Randomization was independent in each hospital. The first paracentesis was done on the 6th day in both groups. Previously, after an overnight fast and 2 h of bed rest, blood samples were drawn to measure plasma renin activity (PRA) in the first 15 patients of each group. An antecubital vein was catheterized and 45 min later, the samples were collected under ice in potassium ethylenediaminetetraacetate (EDTA) tubes, centrifuged at 4°C and the plasma frozen at -30°C until assayed. Plasma renin activity was estimated by the radioimmunoassay kit (CIS Bioirdustries, France). Factor VIII was measured in 12 patients of each group admitted in Hospital Posadas. Serum urea, creatinine, electrolytes, bilirubin and albumin and prothrombin time, platelet count and KPTT were measured in all patients. Daily clinical evaluation was also done, including mean arterial pressure (MAP), heart rate, body weight and urinary volume. Paracentesis was performed under strict sterile conditions with local anesthesia in the left lower abdominal quadrant with an Abbocath 14 G or Kuss needle, with up to 5 liters of ascitic fluid removed in each tap. Samples of ascitic fluid were obtained for polymorpbonuclear cell count and culture in each paracentesis. Dextran 70 containing 6 g Dextran 70 per 100 ml saline isotonic solution (Laboratorios Roux-Ocefa, Buenos Aires, Argentina) and Alburnina Humana Inmuno 20% (Inmuno AG, Vienna, Austria) were used as expanders. Six grams of Cextran and albumin were administered for each 1000 ml of ascites removed from the patients of groups I and II, rzspectively. Paracentesis with expansor infusion was repeated

311 every day until all ascites was removed or when ascitic fluid flow from the cannula became intermittent in spite of mobilization and placing the patient in left lateral decubitus. The morning after the last paracentesis, clinical and laboratory data similar to basal were obtained. In 12 pati-nts of each group the renal and hepatic function test, electrolytes and PRA were repeated 96 h after the last paracentesis and the results are given in Table 3. Patients were given diuretics (spironolactone and furosemide) on the 5th day after paracentesis to prevent recurrence of ascites. Patients were discharged from the hospital and followed closely in the outpatient clinic. Patients who developed tense ascites during the follow-up were readmitted to the hospital and treated according to the initial schedule. Renal impairment was considered present when serum creatinine and serum urea increased > 50% above 1.5 mg/dl or above 40 mgldl, respectively. Hyponatremia was considered present when serum sodium dropped more than 5 mEq below 130 mEq/l, and hyperkalemia if serum potassium increased > 1.5 mEq/l to a level above 5.5 mEqil. The analy& of the results was performed by the Jtest and paired a,rd unpaired Student t-test. Probability curves were consrucred by the Kaplan-Meier method and compared with the Log Rank Test. The results are expressed as mean + S.D.

There were no significant differences between the two groups with respect to clinical data, liver and renal function tests, serum and urine electrolytes, and PRA at the entry into the study (Table 1). Both groups were also similar in the number of patients with previous episodes of ascites (13 patients from group I and 11 patients from group II) and the presence of peripheral edema (14 patients from group I and 13 patients from group II). The subgroups of patients analyzed 96 h after the iast paracentesis were similar to the whole group of patients for clinical and laboratory data and were comparable ir. pretreatment state (Table 2). Ascites disappeared, on physical examination, in 19 of 20 patients from group I (95%) and in all patients from group II (100%). In one patient from group I there was only partial mobilization due to compartmentalization of ascites. There were no significant differences between groups I and II in the volume of ascites removed (12.9 f 4.4 and 10.9 + 3.7 liters, respectively) or the decrease in body weight after treatment (10.4 + 3.4 and 9.6 f 3.6 kg, respectively). Table 3 shows the clinical and biochemical results cf

E. PASS10 et al.

312 both groups of patients before, 24 and 96 h after paracentesis. Serum albumin increased in group II 24 h after treatment (from 2.7 f 0.4 to 3.0 + 0.4 gldl (p < 0.05), but 96 h later the concentration was similar to basal values (2.7 f 0.2 vs 2.7 + 0.4 gldl, respectively). Serum bilirubin decreased from 3.1 + 2.2 to 1.8 k 1.2 mg/dl at 96 h in group I (p c 0.05). There were’ no significant changes in mean arterial pressure (MAP), heart rate, renal function test, serum electrolytes, hematocrit and PRA in either of the groups. A slight but not statistically significant decrease in MAP was observed in both groups at the end of treatment. There were no significant changes in coagulation tests in either groups after treatment (Table 4). Four patients from group I (20%) and four patients from group II (19%) developed complications during their hospital stay (Table 5). Hyponatremia was the most frequent complication (three patients in group I and four in group II). In patients from group I, serum sodium declined from 136, 131 and 125 mEq/l to 129, 123 and 119 mEq/I, respectively. Patients from group II developed hyponatremia, and serum sodium fell from 136, 133, 128 and 125 mEq/l to 128, 122, 122 and 118 mEq/l, respectively. All the patients wer, asymptomatic and recovered spontaneously within 15 days. Renal impairment was observed in one patient from each group. In the patient from group I, serum urea increased from 22 to 30 mg/dl at 24 h after paracentesis and to 50 mgldl at 96 h. In the patient from group II, developing renal impairment serum urea rose from 25 to 37 mg/dl at

TABLE 1

Clinicaland laboratory data at inclusion in the study for patients from group I (treated with P + Dextran) and group II (treated with P + albumin)”

Age (yr) Sex M/P (n) Alcoholism (n) Peripheral edema (n) Child-Pugh (score) MAP (mmHg) Serum bilirubin (mg/dl) Prothrombin (%) Serum albumin (g/dl) Serum urea (mg/dl) Serum creatinine (mg/dl) Serum sodium (mEq/l) Serum potassium (mEq/l) Hematocrit (%) Urine sodium (mEq/d) PRA (ng/ml/h)b

Group I

Group II

(n =

(n = 21)

20)

54 & 7 1416 15 14 10.3 rb 1.1 82 * 8 3.1 + 2.2 61 f 16 2.6 f 0.5 27 k 12 1.0 zk 0.3 132 + 6 4.2 f 0.6 36 i 3 7+4 6.9 + 5

54 + 8 1714 19 13 9.8 + 1.2 82 + 9 2.2 + 2 60 f 17 2.7 F 0.4 29 f 13 1.0 f 0.3 132 k 4 4.2 + 0.5 35 + 4 8f5 7.7 * 5

“None of the values in group I were significantly different from those in group II. bMeasured in 15 patients from each group. P, paracentesis; MAP, mean arterial pressure; PRA, plasma renin activity. Normal values 0.2-2.8 @ml/h.

24 h and to ?2 mg!dl at 96 h after paracentesis. In both cases serum creatinine did not change significantly and both patients recovered the normal renal function test during the follow-up. One patient from group I with encephalopathy died during hospitalization 10 days after the end of the treatment.

TABLE 2 Basal clinical and humoral results for subgroups of patients from group I and group II studied 96 hours after treatment and comparison with basal results from whole group I and whole group Ha Group Ib Subgroup (n = 12) Age (yr) Sex M/F (n) Alcoholism (n) Peripheral edema (n) MAP (mmHg) Bilirubin (mg/dl)* Prothrombin (%) Albumin (g/dl)* Urea (mg/dl)* Creatinine (mg/dl)* Sodium (mEq/I)* Potassium (mEq/l)* Hematocrit (%) Urine sodium (mEq/d) PRA (nglmhh)

53 f 6 715 9 6 80 ?r 9 2.4 f 1.9 63 + 17 2.5 + 0.4 27 f 11 1.1 f 0.3 131 f 6 4.3 +- 0.5 35 * 3 7+5 7.9 z!z5

Group IF Whole group (n = 20) 54 -r-7 1416 15 14 42 k 8 3.1 * 2.2 61 f 16 2.6 + 0.5 27 k 12 1.0 * 0.3 132 f 6 4.2 f 0.6 36 + 3 7+4 6.9 f 5

Subgroup (n = 12) 52 + 8 1012 11 7 78 f 7 1.8 + 0.7 58 + 19 2.6 f 0.2 2a + 15 1.1 + 0.3 133 + 3 4.2 f 0.6 35 r 4 7+5 8.4 + 5

Whole group (n = 21) 54 + 8 1714 19 13 82 + 9 2.2 + 2 60 f 17 2.7 f 0.4 29 + 13 1.0 f 0.3 132 + 4 4.2 f 0.5 35 f 4 df 5 7.7 It 5

*Serum concentration. aThere were no significantdifferences between the basal values from subgroup I and subpup II. bNone of the values in subgroup I were significantly different from those in whole group I. CNone of the values in subgroup II were significantly different from those in whole group II. Group I, Paracentesis with Dextran: group II, pkracentesis with albumin. MAP, mean arterial pressure; PRA, plasma renin activity. Normal values 0.2-2.8 @ml/h.

PARACENTESIS

WITH DEXTRAN

IN CIRRHOTIC

PATIENTS

313

T,4BLE 3 Clinic and humoral results before, 24 and 96 hours after treatment with P + albumin)”

for patients from group I (treated with P + Dexcran) and group II (treated

___.

--_ Group I (n = 20) 24 h

Basal Heart rate (bpm) MAP (mmHg) Bilirubin (mg/dl)* Albumin (g/di)* Urea (mg/dl)* Creatinine (mgldl)* Sodium (r&q/l)* Potassiurn (mEq/i)* Hematocrit (%) PRA (ng/muh)b

81 82 3.1 2.6 27 1.0 132 4.2 36 6.9

+ 10 + 8 e 2.2 + 0.5 + 12 + 0.3 a 6 z!z0.6 + 3 + 5

Group II (n = 21) -~_____ Basal 24 h

95 h’

79& 10 76 f 10 2.2 f 2 2.6 Ifi 0.5 27 z!z 10 1.0 & 0.3 131 * 6 4.1 f 0.3 36 f 3 8.2 & 5

79+ 11 76 + 8 1.8 + 1.2b 2.5 + 0.5 28 + 12 1.0 & 0.2 131 i 8 4.4 f 0.6 35 f 3 8.7 f 4

81 82 2.2 2.7 29 1.0 132 4.2 35 7.7

e r f + t r f + c f

79 80 2.4 3.0 28 1.0 131 4.2 36 9.1

12 9 2 0.4 13 0.3 4 0.5 4 5

+ f -c * & f f k + f

96 h‘

1’ 12 2.1 0.4b 14 0.4 7 0.5 4 9

77 & 9 76 * 10 1.8 t 0.8 2.7 + 0.2 34 f 23 1.1 -+ 0.2 133 c 5 4.0 -c- 0.8 35 f 5 7.9 r 7

*Serum concentration. ‘None of the values post-paracentesis was significantly different from those pre-paracentesis, except in (b). “p c 0.05. Measured patients from each group. dMeasured in 15 patients from each group. Normal values 0.2-2.8 ng/ml/h. P, paracentesis; MAP, mean arterial pressure; PRA. plasma rerun activi5.

in 12

TABLE 4 Coagulation test before, 24 and 96 hours after treatment + albumin)a

~

for patients from group I (treated with P + Dextran) and group II (treated with P

~~

Group I (n = 2G) 24 h

Basal Prothrombin (%) KPI-r (s) Platelets (X 103/ml) Factor VIII (%)’

61 55 137.5 99

+ r + +

16 8 50 14

62 58 126.3 89

Group II (n = 21)

_~_____ 96 hb & + + k

i8 10 47 Jl

69 55 133.1 90

24 h

Basal

C 20 & 15 f 43 & 18

60 52 106.8 84

+ + f +

17 15 43 20

61 57 103.3 84

aNone of the values post-paracentesi.s was significantly differen: from those pre-paracentesis. bThese parameters patients from each group. ‘Measured in 12 patients from each group. Norma: values @l-120%. P. paracentesis; KP’IT, kaolin partial thromboplastin time.

The cost of De?ran infusion was 15.48 dollars for each patient (1 g of Dextran = 0.20 dollars) in group I, while the albumin infusion cost was 364.27 dollars for each patient (1 g of albumin = 5.37 dollars). The expansor cost for each 1000 ml of ascites removed was

i + f i

20 15 45 2;’

62 57 105.1 81

+ 2 * +

22 12 51 28

were measured in 12

33.42 dollars for albumin vs. 1.20 doilars for Dextran. One patient trom group I and two from group II were lost to follow-up at 10, 12 and 13 weeks, respectively. The mean follow-up period in patients from group I and in patients from group II was 32.4 k 28 and 26.9 -t 25 ‘“1 -i

ET. 1. Probability of requiring readmission to hospital during follow-up due to tense ascites in patients treated with paracentesis with Dextran and in patients treated with paracentesis with albumin.

96 hh

_._.._ _._. L

; ..........

Fig. 2. Probability of survival after discharge from hospital in patients treated with paracentesis with D:xtran and in patients treated with paracentesis with albumin.

E. FASSIO et al.

314 TABLE 5 Complications during hospital stay in patients from group I (treated with P + dextran) and group II (treated with P + albumin) Group II p Group I (n = 21) (n = 20) Patients with complications No. of complications Hyponatremia Renal impairment Encephalopathy P, paracentesis;

4 6 3 1 2

4 6 4 1 1

NS NS NS NS NS

NS. not significant.

weeks, respectively. Five patients from group I and six from group II were readmitted to the hospital during follow-up because of recurrence of tense ascites and were treated in the same way as in the first admission. No complications were observed in these patients after paracentesis. The probability of readmission because of ascites was almost identical in both groups of patients (Fig. 1). During follow-up six patients of each group died. In group I the causes of death were: hepatic failure in two patients, sepsis in two, hepatorenal syndrome in one and gastric cancer in one. In group II, gastrointestinal bleeding was the cause of death in three patients, hepatic failure in one, sepsis in one and hepatorenal syndrome in one patient. There were no differences in the probability of survival between patients treated with Dtixtran or albumin (Fig. 2).

Discussion This controlled study shows that Dextran i ‘) can be associated with paracentesis and used as a saft and effective plasma expander in the treatment of cirrhotic patients with tense ascites. We recently demdnstrated that the administration of 100 ml of 6% De>.tran 70 for each 1000 ml of ascites avoids the fall of pulmonary capillary wedge pressure and cardiac output, observed 12 h after up to 5 1 of paracentesis (5). Moreover, in this uncontrolled study of 20 cirrhotic patients, no significant changes were observed in hepatic and renal function tests, the complications related to paracentesis were less than 20% and no patient developed renal impairment. The result of this study supported other studies which have shown the necessity for administering a plasma expander during large volume paracentesis (6,7). Ginf:s et al. demonstrated that paracentesis without albumin w,as associated with renal failure, electrolyte complications and the activation of endogenous vasoactive systems. In contrst, therapeutic paracentesis associated with i.v. al-

bumin infusion did not produce significant changes in vasoactive systems or renal impairment (4). Despite its effectiveness, albumin is not extens.gely used in our country due to its high cost. The results of the current study indicate that paracentesis with Dextran 70 is an effective and safe treatment for patients with cirrhosis and ascites due to the low incidence of complications (20%). The main complication was hyponatremia, and patients recovered spontaneously, without significant changes in hepatic and renal function test. During hospitalization the results in the Dextran group were similar to the albumin group. Moreover, the clinical course after discharge from the hospital was similar in both groups as estimated by the number of patients readmitted by ascites and the probability of surviva1. Our study showed a higher incidence of hyponatremia in both groups than found in two previous studies with similar volume paracentesis (1,2). Although the incidence of hyponatremia in those studies using albumin was less than 5%, in a recent controlled study comparing total paracentesis with Hemaccel vs. total paracentesis with albumin, the incidence of hyponatremia was 15 and 18.5%, respectively (8). We are unable to explain this difference. Nevertheless, in three out of the seven patients who developed hyponatremia, the mean value of serum sodium was 124 + 4 mEq/l which might indicate that they are more prone to develop hyponatremia. It has been suggested that paracentesis without expansion in&ices rapid reaccumulation of ascites, a drop in blood volume followed by renal failure and dilutional hyponatremia, especially in patients without peripheral edema (9,101. However, in our study there was no association of hyponatremia and hepatic failure and/or increase in PRA, and patients spontaneously recovered from this complication during follow-up. Recently, Planes et al. performed a randomized study using Dextran 70 vs. albumin as a plasma expander in cirrhotic patients with tense ascites treated with total paracentesis (11). In this study there was a significant increase ifi PRA on the 6th day after paracentesis in 51% of the pati&s treated with Dextran compared to 15% of the patients treated with albumin. The authors suggest that this rise is compatible with hypovolemia and may be due to a more prolonged elimination half-life for human albumin, than for Dextran (12). However, the clinical significance of this increase is impressive since the incidence of complications such as unknown and renal failure and their clinical course were similar in both group’ of patients in the study of Planas et al. Dextrans are high molecular weight polysaccharider composed of glucose residues produced by the enzyme

PARACENTESIS

WITH DEXTRAN

IN CIRRHOTIC

PATIENTS

dextran sucrase during t!re growth of the bacterium Leucono~toc in media containing sucrose. Dextran 70 is a solution containing a mixture of different fractions with molecular weights ranging between 15 000-160 000 (13). The half-life of elimination of Dextran 70 varies between 23-25.5 h (14). These values represent the first phase ot rapid elimination of fractions with lower molecular weight whoq size is well below that of the renal threshold. The oveiali survival of Dextran 70 in the blood is about 4-6 weeks (15). The pharmacokinetics of Dextran 70 have been studied in nonnovolaernic healthy subjects and in patients with normal glomerular filtration rates (16). Thus, the colloid osmotic effect Jf fractions with higher molecular weights, which remain in the blood longer, are unknown. It is also not known whether patients with cirrhosis and assites present altered Dextran pharmacokinetics. In support of our preliminary study, \ve infused a quantity of Dextran 70 which was equivalent to the dose of albumin administered (6 g/l ascites remov+l). It should be noted, however, that the colloid 0:.motic effect of Dextran (maximum capacity to retain water) is 29 ml of water/g of infused colloid vs. 1.5ml of wat :ug of albumin infused (15,17). In our study, 96 h after paracentesis, 27% of the patients treated with Dextran 70 and 20% of the patients treated with albumin presented an increase of PRA of greater than 30%. However, there are several differences between the study of Planas et al. and our study: (i) Planas et al. treated patients with total paracentesis (9.5 1), while we

315 treated them with repeated paracentesis of up to 5 1. (ii) PRA was determined after a period of 96 h in our study and after 144 h in the study of Planas et al. (iii) in our study, fewer patients were treated. (iv) The quantity of Dextran infused in our study was 6 g/l, whereas in the study of Planas et al. it was 8 g/l. A decrease in factor VIII and platelet alteration have been found in patients receiving Dextran 70, especialIy in doses above l-l.5 g/kg per day (15,18,19). Our patients weie treated with lower doses, and despite repeated paracentesis, there were no accumulative effects demonstrated. Coagulation tests (prothrombin time, platelet count, factor VIII and KPTT) showed no changes after paracentesis with Dextran and were similar to the group treated with albumin. Therefore, we did not find adverse haemostatic effects in patients treated with Dextran. One of the most important observatiorP in the current study is the reduction in cost using Dextran 70 vs. albumin. In this study we confirmed our previous work which suggested that repeated large volume paracentesis associated with Dextran 70 may be the treatment of choice for cirrhotic patients with ascites due to the low cost, effectiveness and safety.

The authors thank Dr. Daniel Flores for his statistical analysis and Mrs. Ana Ras for correcting the English version. Eduardo Fassio was recipient of a grant from Fundacion Profesor Luis Guemes.

eferences Gines P, Arroyo V, Quintero paracentesis and diuretics in

E, et al. Corn-arison

between

9

the treatment of cirrhotrcs with tense ascites. Gastroenterology 1587;93: 234-41. Salerno F, Badalamenti S, Incerti P, et al. Repeated paracentesis and i.v. albumin infusion to treat ‘tense’ ascites in cirrhotic patients: a safe alternative therapy. J H.. ,;atol 1987; 5: 102-8. Tito LI, Gines P, Arroyo V, et al. Total paracentesis associated with intravenous albumin in the management of patients with cirrhosis and ascites. Gastroenterology 1990; 98: 146-51. Gines P, Tito Ll, Arroyo V, et al. Randomized comparative study of therapeutic paracentesis with and without intravenous albumin in cirrhosis. Gastroenterology 1988: 94: 1493-502. Terg R, Romero G. Berreta J, Abecasis R, Boerr L. Dextran administration avoids hemodynamic changes following paracentesis in cirrhotic patients. A safe and inexpensive option. Dig Dis Sci 1991; in press. Simon D, MC Cain J, Bonkovsky H, et al. Effects of therapeutic paracentesis on sy:;temic and hepatic hemodynamics and on renal and hormonal function. Hepatology 1987; 7: 423-9. Panos M, Moore K. Vlaviaons P, et al. Single, total paracentesis for tense ascites: sequential hemodynamics changes and right atria1 size. Hepatology 1990: 11: 662-7. Salerno F, Badalamenti S. Lorenzano E. Moser P, Incerti P. Randomized comparative study of Haemaccel vs. albumin infu-

10 I1

I2

13

14

15

16

sion after total paracentesis in cirrhotic patients with refractory ascites. Hepatology 1991; 13: 707-13. Gabuzda G, Treager H, Davidson C. Hepatic cirrhosis. Factors contributing to the failure to excrete urinary sodium during the acumulation of ascites and edema. J Clin Invest 1950: 20: 81420. Nelson W, Rosembaum J, Stauss M. Hyponatremia in hepatic cirrhosis following paracentesis. J Clin invest 1951; 30: 738-44 Planas R, Gines P, Arroyo V. et al. Dtxtran 70 vs albumin as plasma expanders in cirrhotic patients w!th tense as&es treated with total paracentesis. Results of a randomized study. Gastroenterology 1990; 99: 1736-44. Wilkinson P. Mendenhall C. Serum albumin turnover in nOma sllbjects and patients with cirrhosis measured by 131-I labelled human albumin. Clin Sci 1963; 25: 281-92. KJotx JJ, Kroemer H. Clinical pharmacokine’.ic considerations in the use of plasma expanders. Clin Pharmacokinet 1987: 12: !?3-35. Messmer K. Plasma substitutes. In: Tinker J, Rapin IM. eds. Care of the Critically Ill Patient. New York: Springer-Verlag. 1983; 569-75. Misbler J. Synthetic plasma volume expanders - Their pharmacologv, safety and clinical efficacy. Clin Haematol 1984; 13: 75-92. Arturson Q, Wallenius G. The intravascular perSistenCe Of Dextrm of different molecular sizes in normal humans. Stand J Ctin

316 Lab Invest 1964: 76-80. 17 Barnhart E., ed. Physician’s Desk Reference. 45 Edn. Oradel, NY, Officer of Medical Economic 1991; 599. 18 Korttila K, Groehn P, Gordin A, ~.t al. Effects of hydroxyethyl starch and Dextran on plasma volume and blood hemostasis and

E. FASSIO et al.

coagulation. 3 Clin Pharmacol 1984; 24: 273-82. 19 Aberg M, Hedner U, Bergentz S. Effect of dextran on factor VIII (antihemophilic factor) and platelet function. Ann Surg 1979; 189: 243-7.