GASTROENTEROLOGY
1990;98:146-151
Total Paracentesis Associated With Intravenous Albumin Management of Patients With Cirrhosis and Ascites LLrjCIA TIT(j), PERE GIN&, VICENTE ARROYO, RAMbN PLANAS, JULIA PAN&, ANTON1 RIMOLA, JOSEP LLACH, PERE HUMBERT, SALVATORE BADALAMENTI, WLADIMIRO JIMENEZ, and JOAN ROD& Liver Unit and Hormonal Laboratory. Hospital Clinic i Provincial of Barcelona, Hospital Germans Trias i Pujol of Badalona, Hospital Mtitua of Terrassa, Catalunya, Spain
Repeated large-volume paracentesis (4-6 L/day) is an effective and safe therapy of ascites in patients with cirrhosis provided albumin is infused intravenously. To investigate whether ascites can he safely mobilized in only one paracentesis session (“total paracentesis”), 38 cirrhotic patients with tense ascites were treated with total paracentesis plus intravenous albumin (6-8 g/L ascites removed). Standard liver tests and renal function tests, glomerular filtration rate, free water clearance, plasma volume, plasma renin activity, and plasma aldosterone and norepinephrine concentrations were measured before and after treatment. Total paracentesis was effective in mobilizing ascites in all but 1 patient and did not impair any of the parameters studied. The volume of ascitic fluid removed and the duration of the procedure were lo.7 * 0.5 L (mean + SEM) and 60 * 3 min, respectively. Five of the 38 patients (13%) developed complications during the first hospital stay (hepatic encephalopathy and gastrointestinal hemorrhage in 2 patients each and culture-negative bacterial peritonitis in 1). No patient developed renal impairment. This complication rate, as well as the clinical course of the disease during follow-up, estimated by the probability of readmission to hospital, causes of readmission, and survival probability after treatment, was similar to that reported in patients treated with repeated large-volume paracentesis. These results indicate that total paracentesis associated with intravenous albumin can be safely performed in cirrhotic patients with tense ascites and suggest that these patients could be treated in a single-day hospitalization regime.
P
revious investigations have shown that repeated large-volume paracentesis (4-6 L/day) until complete mobilization of the intraabdominal fluid plus iv. albumin (20-60 g/day] is a fast, effective, and safe therapy of ascites in patients with cirrhosis (l-3), and that the i.v. administration of albumin is essential to prevent the impairment of systemic hemodynamics and renal function that frequently occur when repeated large-volume paracentesis is performed without intravascular volume expansion (4). The current study was undertaken to investigate whether tense ascites in cirrhosis can be mobilized in only one paracentesis session without adverse effects on systemic hemodynamics and renal function. This “total paracentesis,” if safe, would considerably simplify the treatment of these patients. Patients and Methods This study includes 38 cirrhotic patients admitted for the treatment of an episode of ascites in three hospitals from the Barcelona area (Hospital Clinic i Provincial of Barcelona, Hospital Germans Trias i Pujol of Badalona, and Hospital Mlitua of Terrassa) between January 1987 and January 1988. The criteria for admitting patients into the study were (a) cirrhosis with tense ascites (in 29 patients the diagnosis of cirrhosis was based on liver biopsy and in the remaining 9 patients on the following clinical and laboratory data and exploratory findings: signs of chronic liver disease,
Abbreviations used in this paper: BUN, blood urea nitrogen; PRA, plasma renin activity; NE, norepinephrine; GPR,glomerular filtration rate; PAC, plasma aldosterone concentration. 0 1990by the American Gastroenterological Association 0016-5085/90/$3.00
January 1990
abnormal 99mTc-liver-spleen scan, prolonged prothrombin time, hypoalbuminemia, hypergammaglobulinemia, and esophageal varices at endoscopy or collateral circulation at ecography); (b) no clinical, biochemical, or echographic data suggesting hepatocellular carcinoma; [c] absence of hepatic encephalopathy, gastrointestinal hemorrhage, or infection at inclusion; Id] serum bilirubin t10 mg/dl; (e) prothrombin time ~3.5 s over control (>40%); (f) platelet count >40,000/ mm3; (g) serum creatinine ~3 mg/dl; and (h) urinary sodium excretion t10 mEq/day. The study was approved by the Investigation and Ethics Committee of each hospital involved, and all patients gave informed consent to participate in the study. After admission, patients were given a diet containing 50 mmol/day of sodium and did not receive diuretics. Water ingestion was restricted (~500 ml/day] in those patients with hyponatremia (serum sodium tl30 mEq/L). On the fifth day, blood samples were taken to measure serum electrolytes, blood urea nitrogen, (BUN] serum creatinine, and liver tests, and urine was collected for 24 h to measure urine electrolytes. On the morning of the sixth day, after overnight fasting from solid food and after 2 h of bed rest, an antecubital vein was catheterized. Forty-five minutes later blood samples were taken to measure plasma renin activity and plasma aldosterone and norepinephrine concentrations. Glomerular filtration rate (GFR) (inulin clearance) and free water clearance after a water overload of 20 ml/kg body wt were also measured in 28 of these patients. In 15 patients, plasma volume was determined using ‘ZSI-labeled human serum albumin. Methods used for these investigations have been described in detail elsewhere (5). After these measurements patients were treated with total paracentesis plus i.v. albumin infusion (Seroalbtimina Humana Hubber 20%; sodium concentration 30 mEq/L; potassium concentration 0.4 mEq/ L). Six to eight grams of albumin per liter of ascites removed was given at a rate of 2 ml/min. To avoid acute intravascular fluid overload, 50% of the albumin was infused immediately after paracentesis and 50% 6 h later. Paracentesis was performed as previously described (1,2). Briefly, under strict sterile conditions and after local anesthesia, a modified Kuss needle [a sharp-pointed blind metal needle within a 7-cmlong 17-gauge metal blunt-edged cannula with sideholes) was inserted in the left lower abdominal quadrant. Once the needle entered the peritoneal cavity, the inner part was removed and the ascitic fluid was mobilized with the aid of a large-volume capacity (6 L) suction pump (Ordisi S.A., Barcelona, Spain). The physician remained at the patient’s bedside during the entire procedure. Paracentesis was finished when ascitic fluid flow from the cannula became intermittent in spite of gentle mobilization of the cannula within the abdominal cavity and turning the patient to his left side. After the procedure patients reclined for 2 h on the side opposite the paracentesis side to prevent leakage of ascitic fluid. Samples of ascitic fluid were routinely taken for cell count, biochemical examination, and culture. After treatment, patients remained in the hospital for 6 days without diuretics. Serum and urine electrolytes, standard renal function tests, liver tests, plasma renin activity, and plasma aldosterone and norepinephrine concentrations were measured 48 h and 6 days after treatment in all
TOTAL PARACENTESIS IN CIRRHOSIS
147
patients. Measurements of glomerular filtration rate and free water clearance (and plasma volume) were repeated 48 h after paracentesis in the 28 (and 15) patients in whom these parameters were determined before treatment. Patients were discharged from the hospital with diuretics to prevent recurrence of ascites and were followed up closely in the outpatient clinic. Initially, patients without renal failure were given 40 mg/day of furosemide and 200 mg/day of spironolactone and those with renal failure were given 80 mg/day of furosemide and 300 mg/day spironolactone. Diuretic dosage was subsequently adjusted to individual responses. Disturbances in renal function and serum electrolytes occurring during hospitalization were defined as follows: [a) renal impairment: 50% increase in serum creatinine or BUN or both, to a level >1.5 mg/dl and 30 mg/dl, respectively, after treatment; (b) hyponatremia: decrease in serum sodium >5 mEq/L to a level ~130 mEq/L; and (c) hyperkalemia when serum potassium increased >1.5 mEq/L to a level >5.5 mEq/L. To simplify the presentation of the results, patients with a serum sodium concentration of ~130 mEq/L before treatment who showed a decrease in serum sodium of >5 mEq/L during hospitalization were also included in the group of patients developing hyponatremia. The analysis of the results was performed in June 1988 by using the paired Student’s t-test, the analysis of variance, and the Newman-Keuls test. Probability curves were constructed by the Kaplan and Meier method. Results are given asmean + SEM.
Results Characteristics
of the Patients
Of the 38 patients included, 33 were men and 5 were women. The mean age was 55.3 + 1.9 yr (range 35-75 yr). The etiology of cirrhosis was alcoholic in 29 patients, cryptogenic in 7, and HBsAg-associated in 2. The liver tests and renal function tests of the patients included are shown in Table 1. Most patients had advanced liver disease as indicated by the high frequency of positive history of complications (i.e., ascites, encephalopathy, gastrointestinal hemorrhage) and a marked alteration of liver tests. Twenty-nine patients (76%) had had at least one previous episode of ascites, 9 (24%) of encephalopathy, and 9 (24%) of gastrointestinal bleeding. Serum bilirubin levels were increased (21 mg/dl] and prothrombin time was prolonged (>1.5 s over control] in 34 (89%) and 33 (87%] patients, respectively. Serum albumin concentration was reduced (~35 g/L) in 35 (92%) patients. At inclusion into the study 6 patients had renal failure (mean serum creatinine and BUN: 1.9 + 0.3 mg/dl and 37 k 3 mg/dl, respectively], 7 had hyponatremia (serum sodium < 130 mEq/L; mean value 126 + 2 mEq/L), and 26 had peripheral edema. The total protein concentration in ascitic fluid was tl g/d1 in 27 patients.
GASTROENTEROLOGY
148 TIT6 ET AL.
Table 1. Liver-and Renal Function Tests, Mean Arterial Pressure, Plasma Volume, Plasma Renin Activity, and Plasma Aldosterone and Norepinephrine Concentrations Before and After Total Paracentesis and intravenous Albumin Before
48 h after
6 days after
Serum bilirubin
ImddJl
2.8 + 0.2
2.7 + 0.2
2.7 YL0.2
28.8 f 0.7
31.0 f 0.9
30.5 f 0.8
2.7 + 0.3 20 f 2
2.7 + 0.3 18 k 2
2.6 + 0.2 19 * 2
1.1 f 0.1
1.0 f 0.1
1.0 + 0.1
136 i 1
135 f 1
135 + 1
4.3 f 0.1
4.3 f 0.1
4.3 f 0.1
282 + 2
281 + 2
281 + 2
2.0 f 0.4
3.0 f 0.9
2.7 + 0.7
93 8.2 167 890 72
85 8.9 155 791 80
89 + 2b 7.8 f 1.5 173 * 48 751 _+89 -
Vol. 98. No. 1
LITERS
25 20
Serum albumin (g/
Jl Prothrombin (seconds over control) BUN (mg/dJ) Serum creatinine
lmg/Ll
15
Serum sodium
@W/L1
10
Serum potassium
@-W/L) Plasma osmolality (mosmoVkg) Sodium excretion
@JWdayl Mean arterial pressure [mmHg) PRA [ng/mJ . h) PAC (ng/dJ)
NE WmJl GFR (mumin) Free water clearance (mUmin) Plasma volume
(mJ1
* + + f t
2 1.6 38 110 7
t f f f +
2’ 1.8 38 124 7
2.5 f 0.6
3.1 f 0.5d
-
4272 k 289
4240 i 286
-
Normal values in our laboratory are as follows: PRA, 1.3 + 0.2 ng/ ml . h; PAC. 10.5 + 1.1 ng/dl; NE, 275 + 46 pg/mk GFR, 100 f 4 ml/min; free water clearance, 9.6 + 0.8 ml/min. “p =z 0.001, bp < 0.005 vs. before paracentesis and albumin administration (analysis of variance and Newman-Keuls test). “Measured in 28 patients. dp < 0.05 (paired Student’s t-test). “Measured in 15 patients. Mean arterial pressure: F = 4.19.
Results During the First Hospital Stay Paracentesis eliminated ascites almost completely in 37 of the 38 patients included. In 1 patient ascites was incompletely mobilized, probably because of intraabdominal compartmentalization of ascitic fluid. The mean volume of ascites removed was 10.7 f 0.5 L (range 6.3-22.5 L) [Figure 1). The mean duration of paracentesis and the rate of ascitic fluid removal were 60 + 3 min [range 30-110 min) and 185 + 7 ml/min (range 112-290 ml/min). In 6 patients the ascitic fluid volume left in the abdominal cavity after total paracentesis (mean volume of ascites removed in these patients: 9.6 * 1.0 L; range 8.7-12.6 L), measured by a dilution technique with ‘?Yc-radiolabeled albumin, was 968 f 161 ml [range 525-1575 ml), indicating that with the method used for paracentesis most of the ascitic fluid could be mobilized. Peripheral edema decreased markedly or resolved completely after para-
5 0 Figure I. Liters of ascitic fluid removed in all patients studied.
centesis in all cases. One patient developed a small hematoma of the abdominal wall after paracentesis. No local complications occurred in the remaining patients. Table 1 shows the effect of total paracentesis plus i.v. albumin infusion on liver tests, renal function tests, mean arterial pressure, plasma volume, plasma renin activity, and plasma aldosterone and norepinephrine concentrations in all patients studied. There were no significant changes in serum bilirubin, serum albumin, or prothrombin time. Total paracentesis plus i.v. albumin infusion did not induce significant changes in BUN, serum creatinine, serum electrolytes, glomerular filtration rate, or plasma volume. A slight but significant reduction in mean arterial pressure after treatment was observed in all patients. However, as expected by the magnitude of the change, this reduction in arterial pressure was completely asymptomatic. Plasma renin activity and plasma aldosterone and norepinephrine concentrations measured 48 h and 6 days after treatment were similar to baseline values. Free water clearance increased significantly after treatment. Table 2 shows that total paracentesis plus iv. albumin did not produce significant changes in renal function, plasma renin activity, or plasma aldosterone or norepinephrine concentrations in patients with renal failure or hyponatremia, or in those without peripheral edema. Of the 38 patients included in the study, 5 (13%) developed complications during the first hospital stay
January 1999
Table 2.
TOTAL PARACENTESIS
IN CIRRHOSIS
149
Renal Function Tests, Mean Arterial Pressure, Plasma Renin Activity, and Plasma Aldosterone and Norepinephrine Concentrations Before and After Total Paracentesis and Intravenous Albumin in Patients With Renal Failure, Hyponatremia, or Both, and in Those Without Peripheral Edema” Patients with renal failure and/or hyponatremia (n = 9) Before
BUN [mg/dJ) Serum creatinine (mg/dl] Serum sodium (mEq/L) Serum potassium (mEq/L) Mean arterial pressure (mmHg) PRA (ng/ml . h) PAC (mg/dl) NEWmU GFR (mVmin)b Free water clearance (mVminlb
26 + 5 1.6zt0.2 128 + 1 4.6+ 0.3 86 + 3 15.7* 4.1 242 i 66 1.189I+C 210 37 f 15 1.3 f 1.0
48 h after 27 + 5 1.6f 0.3 128 i 2 4.4f 0.3 81 + 4 16.9+ 4.9 242 zt 61 931 f 221 50 + 14 2.2 * 1.2
Patients without peripheral edema (n = 121
6 days after 26 f 5 1.6e 0.3 129 f 3 4.5f 0.4 83 zt4 11.7f 3.5 187 e 40 854 f 169 -
Before 21 f 3 1.1* 0.1 136 f 1 4.4* 0.2 9lt 3 9.2f 3.2 215 k 85 857 + 179 80 f IO 2.9 f 1.2
48 h after
6 days after
19 f 3 1.1+ 0.1 136 + 1 4.5f 0.2 84 i 3 11.1f 4.4 209 + 83 902 r 225 88 f 13 3.0 * 1.1
19 * 3 1.1* 0.1 135 + 1 4.7+ 0.2 87 f 3 10.6f 3.1 220 f 85 721 + 148 -
BUN, blood urea nitrogen: GFR, glomerular filtration rate: NE, norepinephrine: PAC, plasma aldosterone concentration: PRA. plasma renin activity. “No change in the variables studied was statistically significant by the analysis variance. bMeasured in 5 patients with renal failure or hyponitremia, or both, and in 8 patients with peripheral edema.
and 2 (5%) died. Two patients developed grade I hepatic encephalopathy that rapidly reverted after treatment with lactulose. Two other patients had gastrointestinal bleeding due to variceal rupture in 1 case and to acute erosions of the gastric mucosa in the other. Both patients also developed hepatic encephalopathy after the hemorrhage. One of these patients died. Finally, 1 patient had culture-negative bacterial peritonitis 8 days after paracentesis and died secondarily to this complication. No cause of peritonitis could be identified at autopsy in this patient. No patient developed renal impairment during the first hospital stay. One patient developed a transient hyponatremia (serum sodium before, and 6 and 11 days after treatment: 133,119, and 132 mEq/L, respectively] and 1 a transient hyperkalemia (serum potassium increased from 4.1 to 6.1 mEq/L]. The mean duration of hospital stay in the 38 patients studied was 12.6 + 1.0 days. Results During
Follow-Up
Two patients were lost from follow-up 1 and 4 mo after discharge from hospital. The remaining patients were followed closely until the end of the study or death. The mean follow-up period in the whole series of patients was 32.1 + 3.9 wk. Twentythree patients required a total of 43 new hospital admissions during follow-up (1.8 readmissions per patient). Causes of readmission are shown in Table 3 and the probability of readmission in Figure 2. Twelve of these 23 patients were readmitted on 24 occasions for the treatment of new episodes of ascites [two readmissions per patient). The remaining patients did not develop ascites or any other major complication that required hospitalization. Fourteen patients died during follow-up. The causes of death were liver
failure in 7 cases, gastrointestinal hemorrhage in 4, bacterial infection in 2, and intracerebral hemorrhage in 1. In the whole series of patients the probability of survival after inclusion into the study is illustrated in Figure 2. Discussion At present three controlled trials have been published evaluating repeated large-volume paracentesis in the treatment of tense ascites in patients with cirrhosis [l-4). In the first study, Gin& et al. (2) compared paracentesis (4-6 L/day until disappearance of ascites) plus i.v. albumin (40 g/day] with conventional diuretic treatment in 117 patients with cirrhosis admitted to the hospital for the treatment of an episode of ascites. They found that therapeutic paracentesis was more effective than diuretic treatment in the mobilization of ascites, that it was associated with a significantly lower incidence of hepatic encephalopathy and of renal and electrolyte disturbances, and that it significantly shortened the duration
Table 3. Patients Readmitted to Hospital During Follow-up” and Number and Causes of Readmissions Patients @I Ascites Gastrointestinal hemorrhage Encephalopathy Bacterial infection Other complications TOTAL “Follow-up was 32.1t 3.9wk.
12 5 4 3 3 23
Readmissions (nl 24 7 5 4 3 43
150
TIT6
GASTROENTEROLOGY
ET AL.
loo80: -J _
A
60-
OJ 0
c 5
I
IO
15
I
20
I
25
1 30
35
40
WEEKS 36
28
20
18
14
11
IO
9
9
5
10
15
20
25
30
35 40 WEEKS
34
27
26
23
19
17
16
80c CI _ -I
60-
z g
401
ii
I 20
B
1 0' 0 38
15
Figure 2. Probability of requiring readmission to hospital during follow-up (A) and probability of survival after entry into the study (B). Figures under each graph represent the number of patients at risk at any given period.
of hospitalstay. In this study it was also shown that, contrary to the traditional concept, mobilization of ascites by paracentesis associated with i.v. albumin did not alter renal function or systemic hemodynamits, the latter being estimated either directly (by measuring plasma volume, cardiac output, and peripheral resistance] or indirectly [by measuring plasma renin activity and plasma norepinephrine and antidiuretic hormone concentrations]. Subsequently, Salerno et al. (3) confirmed these findings in a randomized study including 41 patients. Finally, in a second randomized study comparing repeated large-volume paracentesis with and without albumin infusion in 105 cirrhotics, Gin&s et al. (4) further confirmed the effectiveness and safety of repeated large-volume paracentesis plus iv. albumin infusion in the management of cirrhotics with ascites and demonstrated that i.v. albumin infusion is an important measure in preventing deterioration of systemic hemodynamics and renal function after therapeutic paracentesis, as the mobilization of ascites in the group of patients not receiving
Vol. 98, No. 1
albumin was associated with a marked increase in plasma renin activity in practically every patient, and with the development of renal impairment or dilutional hyponatremia, or both, in 20% of the cases. The occurrence of these renal and electrolyte disturbances was associated with a poor survival rate. A recent study by Simon et al. (6) in 13 patients with cirrhosis and ascites has confirmed that large-volume paracentesis without iv. colloid replacement increases the activity of the renin-angiotensin system and impairs renal and systemic hemodynamics. The results of the current study indicate that total paracentesis plus i.v. albumin infusion is an effective and safe therapy of tense ascites in patients with cirrhosis for the following reasons. [a] The procedure was able to remove ascites almost completely in 37 of the 38 patients included. (b) This was not associated with significant changes in plasma volume, plasma renin activity, plasma aldosterone or norepinephrine concentrations, glomerular filtration rate, blood urea nitrogen, serum creatinine concentration, or serum electrolytes and was accompanied by a slight but significant increase in the renal ability to excrete free water. As plasma renin activity and plasma norepinephrine concentration are very sensitive to changes in effective blood volume, these findings indicate that total paracentesis plus i.v. albumin does not impair systemic hemodynamics and renal function. It is important to point out that in the current study diuretics were not given 5 days before and 6 days after paracentesis. Therefore, whether our results could be extrapolated to patients being treated with diuretics is unknown. (c) The percentage of patients developing complications [namely hepatic encephalopathy, gastrointestinal hemorrhage, or severe bacterial infection) or renal and electrolyte disturbances during their first hospital stay (7 cases, 18%) was similar to that reported in patients treated with repeated largevolume paracentesis plus i.v. albumin infusion (17%) (2,4) and considerably lower than that observed in patients treated with diuretics. In our first randomized trial comparing repeated large-volume paracentesis with conventional diuretic treatment, 61% of the patients in the diuretic group developed complications or renal and electrolyte disturbances during their first hospital stay, the most frequent being hyponatremia, hepatic encephalopathy, and renal impairment, which occurred in 30%, 29%, and 27% of the patients, respectively. These percentages are comparable to those reported by Sherlock et al. (7) (al%, 29%. and 34%) and by Strauss et al. (8) (22%. 27%. and 24%) in two series of 112 and 100 patients, respectively, with nonazotemic and normonatremic cirrhosis and ascites treated with a variety of diuretic combinations. In the current study the incidence of hyponatremia, hepatic encephalopathy, and renal impairment was 3’70,IO%,
January 1990
and 0%) respectively. It is important to remark that the criteria used to recruit patients, the hospitals involved, the technique of paracentesis, and the amount of albumin infused per liter of ascites removed in this study were identical to those of the two controlled trials previously reported by our group (2,4). Therefore, the comparison between these three investigations is relevant in the assessment of the safety of total paracentesis in cirrhosis with ascites. (d) Finally, the clinical course of the patients included in the present study, as evaluated by the probability of readmission into the hospital during follow-up, causes of readmission, and probability of survival after inclusion into the study, was also similar to that observed in patients treated with repeated large-volume paracentesis associated with iv. albumin infusion included in our two previous trials (2,4). The probabilities of readmission 20 and 40 wk after discharge from the hospital were 50% and 65% in the current series, and 48% and 64%, and 47% and 58% in the two previous trials, respectively. The probabilities of survival 20 and 40 wk after inclusion in the three trials were 69% and 62%, 73% and 59%) and 74% and 6570, respectively. The results of the current study, as well as those of our two previous trials (2,4), should not be taken as an indication that therapeutic paracentesis is the treatment of choice for every patient with cirrhosis and ascites. Approximately 25% of these patients excrete relatively high amounts of sodium in the urine and their ascites may be lost simply by reducing sodium content in the diet (9). In other cirrhotic patients with ascites, sodium excretion is not high enough to allow the development of a negative sodium balance on a sodium-restricted diet: however, they easily respond when treated with low doses of diuretics (spironolactone alone or associated with a loop diuretic) (9). These two groups of patients are infrequently seen in hospitals, as they are satisfactorily managed with sodium restriction and low diuretic dosage on an outpatient basis. In our experience, most cirrhotic patients admitted to the hospital for the treatment of ascites are ones who have developed tense ascites despite a standard diuretic treatment. It is this type of patient in whom therapeutic paracentesis is especially indicated, as strict sodium restriction and high diuretic dosage is required, therefore predisposing the patient to diureticinduced complications. In conclusion, our results indicate that tense ascites in cirrhosis can be completely mobilized within 1 or 2 h without adverse effects, providing the intravascular volume is expanded with human serum albumin. The
TOTALPARACENTESISIN CIRRHOSIS
151
subsequent administration of diuretics avoids reaccumulation of ascites in those patients responding to these drugs. Therefore, tense ascites in cirrhosis can be treated in a single-day hospitalization regime. Further studies are necessary to confirm our findings and to investigate whether total paracentesis performed periodically compared favorably with peritoneovenous shunt in those patients with cirrhosis whose ascites is refractory to diuretic treatment.
References 1. Quintero
2.
3.
4.
5.
6.
7. 8.
9.
E, Gin& P. Arroyo V, et al. Paracentesis versus diuretics in the treatment of cirrhotics with tense ascites. Lancet 1985+611-2 ,. . Gin&s P, Arroyo V, Quintero E, et al. Comparison between paracentesis and diuretics in the treatment of cirrhotics with tense ascites. Gastroenterology 1987;93:234-41. Salerno F, Badalamenti S, Incerti P, et al. Repeated paracentesis and iv. albumin infusion to treat “tense” ascites in cirrhotic patients: a safe alternative therapy. J Hepatol1987;5:102-8. Gin& P. Tit6 Ll, Arroyo V, et al. Randomized comparative study of therapeutic paracentesis with and without intravenous albumin in cirrhosis. Gastroenterology 1988:94:1493-502. PBrez-Ayuso RM, Arroyo V. Camps J, et al. Evidence that renal prostaglandins are involved in renal water metabolism in cirrhosis. Kidney Int 1984;26:72-80. Simon DM, McCain JR, Bonkovsky HL, Wells JO, Mardle DK, Galambos JT. Effect of therapeutic paracentesis on systemic and hepatic hemodynamics and on renal and hormonal function. Hepatology 1987;7:423-9. Sherlock S, Senewiratne B, Scott A, Walker JG. Complications of diuretic therapy in hepatic cirrhosis. Lancet 1966;i:1049-53. Strauss E, De Sa MF, Lacet CM, et al. Standardization of a therapeutic approach for ascites due to chronic liver disease. A prospective study of 100 cases. GED 1985;4:79-86. Arroyo V, Rod&s J. A rational approach to the treatment of ascites. Postgrad Med J 1975;51:558-62.
Received November 14.1988. Accepted May 19.1989. Address requests for reprints to: Vicente Arroyo, M.D., Unitat d’Hepatologia, Hospital Clinic i Provincial, Villarroel 170, 08036 Barcelona, Spain. This work was supported by grants from the Fondo de Investigaciones Sanitarias de la Seguridad Social (FISS 88/1.167) and from Direcci6n General de Investigacibn Cientifico y TBcnica (PA860405). P. Gin&, Ll. Tit6, and J. Llach were recipients of other grants from the Fondo de Investigaciones Sanitarias de la Seguridad Social. The authors also acknowledge the support of the Fundacid Catalana per a I’Estudi de les Malalties de1 Fe$e. The authors thank Drs. C. Piera. J. Viver, J. Gaya, and Mrs. M. Roman, Mrs. E. Calvo, and Mrs. C. Escofet for participation in the study. Mrs. Eulelia Ventura for secretarial assistance; and the medical and nursing staff of the Liver Unit and Hormonal Laboratory of the Hospital Clinic i Provincial for the enthusiastic collaboration.
1990;98:146-151
Total Paracentesis Associated With Intravenous Albumin Management of Patients With Cirrhosis and Ascites LLrjCIA TIT(j), PERE GIN&, VICENTE ARROYO, RAMbN PLANAS, JULIA PAN&, ANTON1 RIMOLA, JOSEP LLACH, PERE HUMBERT, SALVATORE BADALAMENTI, WLADIMIRO JIMENEZ, and JOAN ROD& Liver Unit and Hormonal Laboratory. Hospital Clinic i Provincial of Barcelona, Hospital Germans Trias i Pujol of Badalona, Hospital Mtitua of Terrassa, Catalunya, Spain
Repeated large-volume paracentesis (4-6 L/day) is an effective and safe therapy of ascites in patients with cirrhosis provided albumin is infused intravenously. To investigate whether ascites can he safely mobilized in only one paracentesis session (“total paracentesis”), 38 cirrhotic patients with tense ascites were treated with total paracentesis plus intravenous albumin (6-8 g/L ascites removed). Standard liver tests and renal function tests, glomerular filtration rate, free water clearance, plasma volume, plasma renin activity, and plasma aldosterone and norepinephrine concentrations were measured before and after treatment. Total paracentesis was effective in mobilizing ascites in all but 1 patient and did not impair any of the parameters studied. The volume of ascitic fluid removed and the duration of the procedure were lo.7 * 0.5 L (mean + SEM) and 60 * 3 min, respectively. Five of the 38 patients (13%) developed complications during the first hospital stay (hepatic encephalopathy and gastrointestinal hemorrhage in 2 patients each and culture-negative bacterial peritonitis in 1). No patient developed renal impairment. This complication rate, as well as the clinical course of the disease during follow-up, estimated by the probability of readmission to hospital, causes of readmission, and survival probability after treatment, was similar to that reported in patients treated with repeated large-volume paracentesis. These results indicate that total paracentesis associated with intravenous albumin can be safely performed in cirrhotic patients with tense ascites and suggest that these patients could be treated in a single-day hospitalization regime.
P
revious investigations have shown that repeated large-volume paracentesis (4-6 L/day) until complete mobilization of the intraabdominal fluid plus iv. albumin (20-60 g/day] is a fast, effective, and safe therapy of ascites in patients with cirrhosis (l-3), and that the i.v. administration of albumin is essential to prevent the impairment of systemic hemodynamics and renal function that frequently occur when repeated large-volume paracentesis is performed without intravascular volume expansion (4). The current study was undertaken to investigate whether tense ascites in cirrhosis can be mobilized in only one paracentesis session without adverse effects on systemic hemodynamics and renal function. This “total paracentesis,” if safe, would considerably simplify the treatment of these patients. Patients and Methods This study includes 38 cirrhotic patients admitted for the treatment of an episode of ascites in three hospitals from the Barcelona area (Hospital Clinic i Provincial of Barcelona, Hospital Germans Trias i Pujol of Badalona, and Hospital Mlitua of Terrassa) between January 1987 and January 1988. The criteria for admitting patients into the study were (a) cirrhosis with tense ascites (in 29 patients the diagnosis of cirrhosis was based on liver biopsy and in the remaining 9 patients on the following clinical and laboratory data and exploratory findings: signs of chronic liver disease,
Abbreviations used in this paper: BUN, blood urea nitrogen; PRA, plasma renin activity; NE, norepinephrine; GPR,glomerular filtration rate; PAC, plasma aldosterone concentration. 0 1990by the American Gastroenterological Association 0016-5085/90/$3.00
January 1990
abnormal 99mTc-liver-spleen scan, prolonged prothrombin time, hypoalbuminemia, hypergammaglobulinemia, and esophageal varices at endoscopy or collateral circulation at ecography); (b) no clinical, biochemical, or echographic data suggesting hepatocellular carcinoma; [c] absence of hepatic encephalopathy, gastrointestinal hemorrhage, or infection at inclusion; Id] serum bilirubin t10 mg/dl; (e) prothrombin time ~3.5 s over control (>40%); (f) platelet count >40,000/ mm3; (g) serum creatinine ~3 mg/dl; and (h) urinary sodium excretion t10 mEq/day. The study was approved by the Investigation and Ethics Committee of each hospital involved, and all patients gave informed consent to participate in the study. After admission, patients were given a diet containing 50 mmol/day of sodium and did not receive diuretics. Water ingestion was restricted (~500 ml/day] in those patients with hyponatremia (serum sodium tl30 mEq/L). On the fifth day, blood samples were taken to measure serum electrolytes, blood urea nitrogen, (BUN] serum creatinine, and liver tests, and urine was collected for 24 h to measure urine electrolytes. On the morning of the sixth day, after overnight fasting from solid food and after 2 h of bed rest, an antecubital vein was catheterized. Forty-five minutes later blood samples were taken to measure plasma renin activity and plasma aldosterone and norepinephrine concentrations. Glomerular filtration rate (GFR) (inulin clearance) and free water clearance after a water overload of 20 ml/kg body wt were also measured in 28 of these patients. In 15 patients, plasma volume was determined using ‘ZSI-labeled human serum albumin. Methods used for these investigations have been described in detail elsewhere (5). After these measurements patients were treated with total paracentesis plus i.v. albumin infusion (Seroalbtimina Humana Hubber 20%; sodium concentration 30 mEq/L; potassium concentration 0.4 mEq/ L). Six to eight grams of albumin per liter of ascites removed was given at a rate of 2 ml/min. To avoid acute intravascular fluid overload, 50% of the albumin was infused immediately after paracentesis and 50% 6 h later. Paracentesis was performed as previously described (1,2). Briefly, under strict sterile conditions and after local anesthesia, a modified Kuss needle [a sharp-pointed blind metal needle within a 7-cmlong 17-gauge metal blunt-edged cannula with sideholes) was inserted in the left lower abdominal quadrant. Once the needle entered the peritoneal cavity, the inner part was removed and the ascitic fluid was mobilized with the aid of a large-volume capacity (6 L) suction pump (Ordisi S.A., Barcelona, Spain). The physician remained at the patient’s bedside during the entire procedure. Paracentesis was finished when ascitic fluid flow from the cannula became intermittent in spite of gentle mobilization of the cannula within the abdominal cavity and turning the patient to his left side. After the procedure patients reclined for 2 h on the side opposite the paracentesis side to prevent leakage of ascitic fluid. Samples of ascitic fluid were routinely taken for cell count, biochemical examination, and culture. After treatment, patients remained in the hospital for 6 days without diuretics. Serum and urine electrolytes, standard renal function tests, liver tests, plasma renin activity, and plasma aldosterone and norepinephrine concentrations were measured 48 h and 6 days after treatment in all
TOTAL PARACENTESIS IN CIRRHOSIS
147
patients. Measurements of glomerular filtration rate and free water clearance (and plasma volume) were repeated 48 h after paracentesis in the 28 (and 15) patients in whom these parameters were determined before treatment. Patients were discharged from the hospital with diuretics to prevent recurrence of ascites and were followed up closely in the outpatient clinic. Initially, patients without renal failure were given 40 mg/day of furosemide and 200 mg/day of spironolactone and those with renal failure were given 80 mg/day of furosemide and 300 mg/day spironolactone. Diuretic dosage was subsequently adjusted to individual responses. Disturbances in renal function and serum electrolytes occurring during hospitalization were defined as follows: [a) renal impairment: 50% increase in serum creatinine or BUN or both, to a level >1.5 mg/dl and 30 mg/dl, respectively, after treatment; (b) hyponatremia: decrease in serum sodium >5 mEq/L to a level ~130 mEq/L; and (c) hyperkalemia when serum potassium increased >1.5 mEq/L to a level >5.5 mEq/L. To simplify the presentation of the results, patients with a serum sodium concentration of ~130 mEq/L before treatment who showed a decrease in serum sodium of >5 mEq/L during hospitalization were also included in the group of patients developing hyponatremia. The analysis of the results was performed in June 1988 by using the paired Student’s t-test, the analysis of variance, and the Newman-Keuls test. Probability curves were constructed by the Kaplan and Meier method. Results are given asmean + SEM.
Results Characteristics
of the Patients
Of the 38 patients included, 33 were men and 5 were women. The mean age was 55.3 + 1.9 yr (range 35-75 yr). The etiology of cirrhosis was alcoholic in 29 patients, cryptogenic in 7, and HBsAg-associated in 2. The liver tests and renal function tests of the patients included are shown in Table 1. Most patients had advanced liver disease as indicated by the high frequency of positive history of complications (i.e., ascites, encephalopathy, gastrointestinal hemorrhage) and a marked alteration of liver tests. Twenty-nine patients (76%) had had at least one previous episode of ascites, 9 (24%) of encephalopathy, and 9 (24%) of gastrointestinal bleeding. Serum bilirubin levels were increased (21 mg/dl] and prothrombin time was prolonged (>1.5 s over control] in 34 (89%) and 33 (87%] patients, respectively. Serum albumin concentration was reduced (~35 g/L) in 35 (92%) patients. At inclusion into the study 6 patients had renal failure (mean serum creatinine and BUN: 1.9 + 0.3 mg/dl and 37 k 3 mg/dl, respectively], 7 had hyponatremia (serum sodium < 130 mEq/L; mean value 126 + 2 mEq/L), and 26 had peripheral edema. The total protein concentration in ascitic fluid was tl g/d1 in 27 patients.
GASTROENTEROLOGY
148 TIT6 ET AL.
Table 1. Liver-and Renal Function Tests, Mean Arterial Pressure, Plasma Volume, Plasma Renin Activity, and Plasma Aldosterone and Norepinephrine Concentrations Before and After Total Paracentesis and intravenous Albumin Before
48 h after
6 days after
Serum bilirubin
ImddJl
2.8 + 0.2
2.7 + 0.2
2.7 YL0.2
28.8 f 0.7
31.0 f 0.9
30.5 f 0.8
2.7 + 0.3 20 f 2
2.7 + 0.3 18 k 2
2.6 + 0.2 19 * 2
1.1 f 0.1
1.0 f 0.1
1.0 + 0.1
136 i 1
135 f 1
135 + 1
4.3 f 0.1
4.3 f 0.1
4.3 f 0.1
282 + 2
281 + 2
281 + 2
2.0 f 0.4
3.0 f 0.9
2.7 + 0.7
93 8.2 167 890 72
85 8.9 155 791 80
89 + 2b 7.8 f 1.5 173 * 48 751 _+89 -
Vol. 98. No. 1
LITERS
25 20
Serum albumin (g/
Jl Prothrombin (seconds over control) BUN (mg/dJ) Serum creatinine
lmg/Ll
15
Serum sodium
@W/L1
10
Serum potassium
@-W/L) Plasma osmolality (mosmoVkg) Sodium excretion
@JWdayl Mean arterial pressure [mmHg) PRA [ng/mJ . h) PAC (ng/dJ)
NE WmJl GFR (mumin) Free water clearance (mUmin) Plasma volume
(mJ1
* + + f t
2 1.6 38 110 7
t f f f +
2’ 1.8 38 124 7
2.5 f 0.6
3.1 f 0.5d
-
4272 k 289
4240 i 286
-
Normal values in our laboratory are as follows: PRA, 1.3 + 0.2 ng/ ml . h; PAC. 10.5 + 1.1 ng/dl; NE, 275 + 46 pg/mk GFR, 100 f 4 ml/min; free water clearance, 9.6 + 0.8 ml/min. “p =z 0.001, bp < 0.005 vs. before paracentesis and albumin administration (analysis of variance and Newman-Keuls test). “Measured in 28 patients. dp < 0.05 (paired Student’s t-test). “Measured in 15 patients. Mean arterial pressure: F = 4.19.
Results During the First Hospital Stay Paracentesis eliminated ascites almost completely in 37 of the 38 patients included. In 1 patient ascites was incompletely mobilized, probably because of intraabdominal compartmentalization of ascitic fluid. The mean volume of ascites removed was 10.7 f 0.5 L (range 6.3-22.5 L) [Figure 1). The mean duration of paracentesis and the rate of ascitic fluid removal were 60 + 3 min [range 30-110 min) and 185 + 7 ml/min (range 112-290 ml/min). In 6 patients the ascitic fluid volume left in the abdominal cavity after total paracentesis (mean volume of ascites removed in these patients: 9.6 * 1.0 L; range 8.7-12.6 L), measured by a dilution technique with ‘?Yc-radiolabeled albumin, was 968 f 161 ml [range 525-1575 ml), indicating that with the method used for paracentesis most of the ascitic fluid could be mobilized. Peripheral edema decreased markedly or resolved completely after para-
5 0 Figure I. Liters of ascitic fluid removed in all patients studied.
centesis in all cases. One patient developed a small hematoma of the abdominal wall after paracentesis. No local complications occurred in the remaining patients. Table 1 shows the effect of total paracentesis plus i.v. albumin infusion on liver tests, renal function tests, mean arterial pressure, plasma volume, plasma renin activity, and plasma aldosterone and norepinephrine concentrations in all patients studied. There were no significant changes in serum bilirubin, serum albumin, or prothrombin time. Total paracentesis plus i.v. albumin infusion did not induce significant changes in BUN, serum creatinine, serum electrolytes, glomerular filtration rate, or plasma volume. A slight but significant reduction in mean arterial pressure after treatment was observed in all patients. However, as expected by the magnitude of the change, this reduction in arterial pressure was completely asymptomatic. Plasma renin activity and plasma aldosterone and norepinephrine concentrations measured 48 h and 6 days after treatment were similar to baseline values. Free water clearance increased significantly after treatment. Table 2 shows that total paracentesis plus iv. albumin did not produce significant changes in renal function, plasma renin activity, or plasma aldosterone or norepinephrine concentrations in patients with renal failure or hyponatremia, or in those without peripheral edema. Of the 38 patients included in the study, 5 (13%) developed complications during the first hospital stay
January 1999
Table 2.
TOTAL PARACENTESIS
IN CIRRHOSIS
149
Renal Function Tests, Mean Arterial Pressure, Plasma Renin Activity, and Plasma Aldosterone and Norepinephrine Concentrations Before and After Total Paracentesis and Intravenous Albumin in Patients With Renal Failure, Hyponatremia, or Both, and in Those Without Peripheral Edema” Patients with renal failure and/or hyponatremia (n = 9) Before
BUN [mg/dJ) Serum creatinine (mg/dl] Serum sodium (mEq/L) Serum potassium (mEq/L) Mean arterial pressure (mmHg) PRA (ng/ml . h) PAC (mg/dl) NEWmU GFR (mVmin)b Free water clearance (mVminlb
26 + 5 1.6zt0.2 128 + 1 4.6+ 0.3 86 + 3 15.7* 4.1 242 i 66 1.189I+C 210 37 f 15 1.3 f 1.0
48 h after 27 + 5 1.6f 0.3 128 i 2 4.4f 0.3 81 + 4 16.9+ 4.9 242 zt 61 931 f 221 50 + 14 2.2 * 1.2
Patients without peripheral edema (n = 121
6 days after 26 f 5 1.6e 0.3 129 f 3 4.5f 0.4 83 zt4 11.7f 3.5 187 e 40 854 f 169 -
Before 21 f 3 1.1* 0.1 136 f 1 4.4* 0.2 9lt 3 9.2f 3.2 215 k 85 857 + 179 80 f IO 2.9 f 1.2
48 h after
6 days after
19 f 3 1.1+ 0.1 136 + 1 4.5f 0.2 84 i 3 11.1f 4.4 209 + 83 902 r 225 88 f 13 3.0 * 1.1
19 * 3 1.1* 0.1 135 + 1 4.7+ 0.2 87 f 3 10.6f 3.1 220 f 85 721 + 148 -
BUN, blood urea nitrogen: GFR, glomerular filtration rate: NE, norepinephrine: PAC, plasma aldosterone concentration: PRA. plasma renin activity. “No change in the variables studied was statistically significant by the analysis variance. bMeasured in 5 patients with renal failure or hyponitremia, or both, and in 8 patients with peripheral edema.
and 2 (5%) died. Two patients developed grade I hepatic encephalopathy that rapidly reverted after treatment with lactulose. Two other patients had gastrointestinal bleeding due to variceal rupture in 1 case and to acute erosions of the gastric mucosa in the other. Both patients also developed hepatic encephalopathy after the hemorrhage. One of these patients died. Finally, 1 patient had culture-negative bacterial peritonitis 8 days after paracentesis and died secondarily to this complication. No cause of peritonitis could be identified at autopsy in this patient. No patient developed renal impairment during the first hospital stay. One patient developed a transient hyponatremia (serum sodium before, and 6 and 11 days after treatment: 133,119, and 132 mEq/L, respectively] and 1 a transient hyperkalemia (serum potassium increased from 4.1 to 6.1 mEq/L]. The mean duration of hospital stay in the 38 patients studied was 12.6 + 1.0 days. Results During
Follow-Up
Two patients were lost from follow-up 1 and 4 mo after discharge from hospital. The remaining patients were followed closely until the end of the study or death. The mean follow-up period in the whole series of patients was 32.1 + 3.9 wk. Twentythree patients required a total of 43 new hospital admissions during follow-up (1.8 readmissions per patient). Causes of readmission are shown in Table 3 and the probability of readmission in Figure 2. Twelve of these 23 patients were readmitted on 24 occasions for the treatment of new episodes of ascites [two readmissions per patient). The remaining patients did not develop ascites or any other major complication that required hospitalization. Fourteen patients died during follow-up. The causes of death were liver
failure in 7 cases, gastrointestinal hemorrhage in 4, bacterial infection in 2, and intracerebral hemorrhage in 1. In the whole series of patients the probability of survival after inclusion into the study is illustrated in Figure 2. Discussion At present three controlled trials have been published evaluating repeated large-volume paracentesis in the treatment of tense ascites in patients with cirrhosis [l-4). In the first study, Gin& et al. (2) compared paracentesis (4-6 L/day until disappearance of ascites) plus i.v. albumin (40 g/day] with conventional diuretic treatment in 117 patients with cirrhosis admitted to the hospital for the treatment of an episode of ascites. They found that therapeutic paracentesis was more effective than diuretic treatment in the mobilization of ascites, that it was associated with a significantly lower incidence of hepatic encephalopathy and of renal and electrolyte disturbances, and that it significantly shortened the duration
Table 3. Patients Readmitted to Hospital During Follow-up” and Number and Causes of Readmissions Patients @I Ascites Gastrointestinal hemorrhage Encephalopathy Bacterial infection Other complications TOTAL “Follow-up was 32.1t 3.9wk.
12 5 4 3 3 23
Readmissions (nl 24 7 5 4 3 43
150
TIT6
GASTROENTEROLOGY
ET AL.
loo80: -J _
A
60-
OJ 0
c 5
I
IO
15
I
20
I
25
1 30
35
40
WEEKS 36
28
20
18
14
11
IO
9
9
5
10
15
20
25
30
35 40 WEEKS
34
27
26
23
19
17
16
80c CI _ -I
60-
z g
401
ii
I 20
B
1 0' 0 38
15
Figure 2. Probability of requiring readmission to hospital during follow-up (A) and probability of survival after entry into the study (B). Figures under each graph represent the number of patients at risk at any given period.
of hospitalstay. In this study it was also shown that, contrary to the traditional concept, mobilization of ascites by paracentesis associated with i.v. albumin did not alter renal function or systemic hemodynamits, the latter being estimated either directly (by measuring plasma volume, cardiac output, and peripheral resistance] or indirectly [by measuring plasma renin activity and plasma norepinephrine and antidiuretic hormone concentrations]. Subsequently, Salerno et al. (3) confirmed these findings in a randomized study including 41 patients. Finally, in a second randomized study comparing repeated large-volume paracentesis with and without albumin infusion in 105 cirrhotics, Gin&s et al. (4) further confirmed the effectiveness and safety of repeated large-volume paracentesis plus iv. albumin infusion in the management of cirrhotics with ascites and demonstrated that i.v. albumin infusion is an important measure in preventing deterioration of systemic hemodynamics and renal function after therapeutic paracentesis, as the mobilization of ascites in the group of patients not receiving
Vol. 98, No. 1
albumin was associated with a marked increase in plasma renin activity in practically every patient, and with the development of renal impairment or dilutional hyponatremia, or both, in 20% of the cases. The occurrence of these renal and electrolyte disturbances was associated with a poor survival rate. A recent study by Simon et al. (6) in 13 patients with cirrhosis and ascites has confirmed that large-volume paracentesis without iv. colloid replacement increases the activity of the renin-angiotensin system and impairs renal and systemic hemodynamics. The results of the current study indicate that total paracentesis plus i.v. albumin infusion is an effective and safe therapy of tense ascites in patients with cirrhosis for the following reasons. [a] The procedure was able to remove ascites almost completely in 37 of the 38 patients included. (b) This was not associated with significant changes in plasma volume, plasma renin activity, plasma aldosterone or norepinephrine concentrations, glomerular filtration rate, blood urea nitrogen, serum creatinine concentration, or serum electrolytes and was accompanied by a slight but significant increase in the renal ability to excrete free water. As plasma renin activity and plasma norepinephrine concentration are very sensitive to changes in effective blood volume, these findings indicate that total paracentesis plus i.v. albumin does not impair systemic hemodynamics and renal function. It is important to point out that in the current study diuretics were not given 5 days before and 6 days after paracentesis. Therefore, whether our results could be extrapolated to patients being treated with diuretics is unknown. (c) The percentage of patients developing complications [namely hepatic encephalopathy, gastrointestinal hemorrhage, or severe bacterial infection) or renal and electrolyte disturbances during their first hospital stay (7 cases, 18%) was similar to that reported in patients treated with repeated largevolume paracentesis plus i.v. albumin infusion (17%) (2,4) and considerably lower than that observed in patients treated with diuretics. In our first randomized trial comparing repeated large-volume paracentesis with conventional diuretic treatment, 61% of the patients in the diuretic group developed complications or renal and electrolyte disturbances during their first hospital stay, the most frequent being hyponatremia, hepatic encephalopathy, and renal impairment, which occurred in 30%, 29%, and 27% of the patients, respectively. These percentages are comparable to those reported by Sherlock et al. (7) (al%, 29%. and 34%) and by Strauss et al. (8) (22%. 27%. and 24%) in two series of 112 and 100 patients, respectively, with nonazotemic and normonatremic cirrhosis and ascites treated with a variety of diuretic combinations. In the current study the incidence of hyponatremia, hepatic encephalopathy, and renal impairment was 3’70,IO%,
January 1990
and 0%) respectively. It is important to remark that the criteria used to recruit patients, the hospitals involved, the technique of paracentesis, and the amount of albumin infused per liter of ascites removed in this study were identical to those of the two controlled trials previously reported by our group (2,4). Therefore, the comparison between these three investigations is relevant in the assessment of the safety of total paracentesis in cirrhosis with ascites. (d) Finally, the clinical course of the patients included in the present study, as evaluated by the probability of readmission into the hospital during follow-up, causes of readmission, and probability of survival after inclusion into the study, was also similar to that observed in patients treated with repeated large-volume paracentesis associated with iv. albumin infusion included in our two previous trials (2,4). The probabilities of readmission 20 and 40 wk after discharge from the hospital were 50% and 65% in the current series, and 48% and 64%, and 47% and 58% in the two previous trials, respectively. The probabilities of survival 20 and 40 wk after inclusion in the three trials were 69% and 62%, 73% and 59%) and 74% and 6570, respectively. The results of the current study, as well as those of our two previous trials (2,4), should not be taken as an indication that therapeutic paracentesis is the treatment of choice for every patient with cirrhosis and ascites. Approximately 25% of these patients excrete relatively high amounts of sodium in the urine and their ascites may be lost simply by reducing sodium content in the diet (9). In other cirrhotic patients with ascites, sodium excretion is not high enough to allow the development of a negative sodium balance on a sodium-restricted diet: however, they easily respond when treated with low doses of diuretics (spironolactone alone or associated with a loop diuretic) (9). These two groups of patients are infrequently seen in hospitals, as they are satisfactorily managed with sodium restriction and low diuretic dosage on an outpatient basis. In our experience, most cirrhotic patients admitted to the hospital for the treatment of ascites are ones who have developed tense ascites despite a standard diuretic treatment. It is this type of patient in whom therapeutic paracentesis is especially indicated, as strict sodium restriction and high diuretic dosage is required, therefore predisposing the patient to diureticinduced complications. In conclusion, our results indicate that tense ascites in cirrhosis can be completely mobilized within 1 or 2 h without adverse effects, providing the intravascular volume is expanded with human serum albumin. The
TOTALPARACENTESISIN CIRRHOSIS
151
subsequent administration of diuretics avoids reaccumulation of ascites in those patients responding to these drugs. Therefore, tense ascites in cirrhosis can be treated in a single-day hospitalization regime. Further studies are necessary to confirm our findings and to investigate whether total paracentesis performed periodically compared favorably with peritoneovenous shunt in those patients with cirrhosis whose ascites is refractory to diuretic treatment.
References 1. Quintero
2.
3.
4.
5.
6.
7. 8.
9.
E, Gin& P. Arroyo V, et al. Paracentesis versus diuretics in the treatment of cirrhotics with tense ascites. Lancet 1985+611-2 ,. . Gin&s P, Arroyo V, Quintero E, et al. Comparison between paracentesis and diuretics in the treatment of cirrhotics with tense ascites. Gastroenterology 1987;93:234-41. Salerno F, Badalamenti S, Incerti P, et al. Repeated paracentesis and iv. albumin infusion to treat “tense” ascites in cirrhotic patients: a safe alternative therapy. J Hepatol1987;5:102-8. Gin& P. Tit6 Ll, Arroyo V, et al. Randomized comparative study of therapeutic paracentesis with and without intravenous albumin in cirrhosis. Gastroenterology 1988:94:1493-502. PBrez-Ayuso RM, Arroyo V. Camps J, et al. Evidence that renal prostaglandins are involved in renal water metabolism in cirrhosis. Kidney Int 1984;26:72-80. Simon DM, McCain JR, Bonkovsky HL, Wells JO, Mardle DK, Galambos JT. Effect of therapeutic paracentesis on systemic and hepatic hemodynamics and on renal and hormonal function. Hepatology 1987;7:423-9. Sherlock S, Senewiratne B, Scott A, Walker JG. Complications of diuretic therapy in hepatic cirrhosis. Lancet 1966;i:1049-53. Strauss E, De Sa MF, Lacet CM, et al. Standardization of a therapeutic approach for ascites due to chronic liver disease. A prospective study of 100 cases. GED 1985;4:79-86. Arroyo V, Rod&s J. A rational approach to the treatment of ascites. Postgrad Med J 1975;51:558-62.
Received November 14.1988. Accepted May 19.1989. Address requests for reprints to: Vicente Arroyo, M.D., Unitat d’Hepatologia, Hospital Clinic i Provincial, Villarroel 170, 08036 Barcelona, Spain. This work was supported by grants from the Fondo de Investigaciones Sanitarias de la Seguridad Social (FISS 88/1.167) and from Direcci6n General de Investigacibn Cientifico y TBcnica (PA860405). P. Gin&, Ll. Tit6, and J. Llach were recipients of other grants from the Fondo de Investigaciones Sanitarias de la Seguridad Social. The authors also acknowledge the support of the Fundacid Catalana per a I’Estudi de les Malalties de1 Fe$e. The authors thank Drs. C. Piera. J. Viver, J. Gaya, and Mrs. M. Roman, Mrs. E. Calvo, and Mrs. C. Escofet for participation in the study. Mrs. Eulelia Ventura for secretarial assistance; and the medical and nursing staff of the Liver Unit and Hormonal Laboratory of the Hospital Clinic i Provincial for the enthusiastic collaboration.
