Large-volume paracentesis is widely used for the treatment of ascites in cirrhosis. Several randomized trials have shown that it is more effective than diuretics, and is associated with a lower rate of complications and reduced length of hospitalization [1,2]. However, the removal of large amounts of ascitic fluid may induce an impairment of circulatory function that has been termed paracentesis-induced circulatory dysfunction (PICD), first described by Ginès et al. . They observed that most patients treated by large-volume paracentesis without associated infusion of plasma expanders developed a circulatory disorder characterized by marked activation of vasoactive systems, namely plasma renin activity (PRA) and plasma aldosterone concentration. Since then, several studies have investigated the mechanisms leading to the development of PICD but, to date, the origin of this disorder is only partially understood.
Why is PICD relevant?
PICD is a prevalent disorder, appearing in up to 80% of cases when paracentesis is performed without subsequent plasma volume expansion . It consists of a persistent deterioration of the circulatory status that is diagnosed through PRA variations . Soon after large-volume paracentesis, important haemodynamic changes affecting the renin–angiotensin axis may develop. Accordingly, the sensitivity of PRA levels for the diagnosis of PICD 2 days after paracentesis is low ; PICD is best defined by an elevation of PRA of more than 50% to >4 ng/ml/h on the fifth to sixth day after paracentesis.
Some authors consider PICD to be just a ‘cosmetic disturbance’ because it is clinically silent. In fact, few patients show serious clinical complications in the short term, but those developing PICD have a tendency to reaccumulate ascitic fluid faster and to develop hyponatraemia and functional renal failure . Moreover, PICD is not spontaneously reversible and patients who develop this disorder show a reduced survival .
Early haemodynamic changes induced by paracentesis
Characteristic haemodynamic changes following paracentesis begin after removal of as little as 250 ml ascitic fluid  and consist of a marked decrease in right atrial and pulmonary pressures, a decrease in mean arterial pressure, an increase in cardiac output and systemic vasodilatation [4,7–11]. Modifications of the neurohumoral profile are also seen immediately after paracentesis, namely a reduction in PRA and aldosterone levels and an increase in atrial natriuretic peptide (ANP) [4,7,10].
There is no question that most of these changes are favourable overall: an increase in cardiac output, systemic vasodilatation and deactivation of the renin–angiotensin–aldosterone axis reduce fluid retention and favour organ perfusion, thereby alleviating two symptoms that negatively characterize cirrhotic patients with ascites. However, these changes are not maintained indefinitely if large-volume paracentesis is not followed by plasma expansion. Indeed, from 3–24 h after paracentesis, a picture consistent with hypovolaemia develops , being characterized by significant reductions in cardiac index, pulmonary pressures and ANP together with significant increases in PRA and plasma aldosterone concentration .
Initially, some authors proposed that PICD could be caused by a decrease in the total circulating blood volume secondary to a rapid reaccumulation of ascites. However, several studies failed to demonstrate a significant reduction in intravascular volume after removal of large volumes of ascitic fluid [4,12]. Also, in a study by Saló et al. , the plasma volume and transvascular escape of albumin (that estimates the net leakage of fluid from the intravascular compartment to the interstitial fluid) were measured in cirrhotic patients with tense ascites before and 48 h after total paracentesis. PICD occurred in the absence of changes in any of these measures.
How can early haemodynamic changes be explained?
Since plasma volume remains unchanged, an accentuation of arterial vasodilatation, which is already present in patients with advanced cirrhosis, has been suggested to explain the early decrease of systemic vascular resistance (SVR) that develops shortly after paracentesis. The question is why the mobilization of large volumes of ascitic fluid can induce arteriolar vasodilatation. Three pathogenic mechanisms may be considered. In this issue of the Journal, Coll et al.  show that the dynamics of paracentesis constitute an important factor that can influence the early decrease of SVR. The authors report that patients in whom a decrease of SVR (as measured by echo-Doppler) after large-volume paracentesis was observed showed higher baseline intra-abdominal pressure (IAP), together with a shorter duration of paracentesis and higher flow rate of ascites extraction. However, the total volume of ascites drained did not differ between patients who developed a decrease in SVR and those who did not. This is not surprising since other factors, such as abdominal wall compliance, can determine the IAP. An interesting finding of the study by Coll et al.  is the significant correlation found between serum values of nitric oxide metabolites and SVR. The authors suggest that an increase in the synthesis of nitric oxide in response to ‘shear stress’ induced by the increased cardiac output that occurs after paracentesis may be intimately related to post-paracentesis arteriolar vasodilatation.
From a different point of view, it has been suggested that changes in SVR occurring after paracentesis can be explained by mechanical factors. Cabrera et al.  designed a special pneumatic girdle to maintain a constant IAP during paracentesis. When IAP was maintained at its original level, no haemodynamic changes were observed but a significant decrease in SVR appeared immediately after girdle deflation. The authors suggested that the abrupt decrease in IAP after paracentesis induces a deleterious effect due to mechanical decompression of an already dilated splanchnic vascular bed in patients with hypo-responsiveness to vasoconstrictors [16–18]. The beneficial haemodynamic effects derived from changes in cava vein pressure after paracentesis initially hide the deleterious effects of abdominal decompression on the splanchnic vascular bed, which only emerge when the former effect is exhausted.
A third hypothesis is that reduction in SVR is a reflex of the stimulation of cardiac-volume receptors inhibiting sympathetic vasoconstrictor tone and renal release of renin via an increase in cardiac volume . The increase in heart volume as a consequence of increased pre-load after paracentesis may also be directly responsible for the increase in ANP seen immediately after paracentesis.
What do we know about the mechanisms of PICD?
We cannot conclude from these data that the dynamics of paracentesis, release of nitric oxide or a reflex cardiogenic mechanism are the cause of PICD appearing 5–6 days after paracentesis, as the above-mentioned studies were designed to investigate only the early decrease in SVR after paracentesis. Only two studies have been designed to assess the mechanisms of PICD. Ruiz-del-Arbol et al.  observed that most haemodynamic changes occurring immediately after completion of paracentesis persisted 6 days after the procedure, despite expansion with dextran 70. Although cardiac index returned to baseline levels at the sixth day, cardiopulmonary pressures and SVR continued to be significantly reduced with respect to pre-treatment values. Because at this time most patients had less ascitic fluid than before paracentesis, the lower cardiopulmonary pressures could be attributed to a lower intrathoracic pressure. The magnitude of the reduction in SVR 6 days after paracentesis was greater in patients developing PICD, and the reduction occurred in spite of a recovery of cardiac index. This finding suggests that other mechanisms, independent of cardiac output, may be implicated in PICD. The authors also observed a significant inverse correlation between changes in SVR and the degree of activity of renin–angiotensin and the sympathetic nervous system: patients with greater increases in PRA and plasma norepinephrine levels had more intense peripheral vasodilatation.
Similarly, Vila et al.  reported an inverse correlation between PRA and plasma aldosterone concentration and SVR calculated by echo-Doppler. Patients developing PICD showed a significant reduction in SVR 6 days after paracentesis, while those who did not develop PICD showed no change in SVR.
Taken together, these observations suggest that PICD is caused by a further arteriolar vasodilatation and that stimulation of the renin–aldosterone and sympathetic nervous systems corresponds to compensatory mechanisms to maintain circulatory homeostasis. Although more studies designed to investigate the mechanism of PICD are needed, a hypothesis to explain PICD may be formulated as follows. Immediately after paracentesis, an effective hypovolaemia due to accentuation of arteriolar vasodilatation occurs. The origin of this vasodilatation is probably multifactorial and includes an abrupt decrease in IAP, a reflex mechanism via the increase in cardiac output and an increased release of nitric oxide, likely to be secondary to shear stress. In response to this vasodilatation, activation of the renin–angiotensin and sympathetic nervous systems takes place. Patients who are able to compensate for this vasodilatation in the first few days after paracentesis will not develop PICD and the levels of PRA will return to baseline. However, PICD will develop in those who are unable to compensate. The degree of hyporesponsiveness to vasoconstrictors could play an important role in this setting.
What can we do to prevent PICD?
No significant haemodynamic or hormonal variations were reported up to 48 h after paracentesis when a low volume of ascitic fluid was evacuated (≤ 5 l) . In addition, the volume of ascites has been related to the incidence of PICD in two studies [5,22]. Therefore, reducing the volume of ascites drained seems an effective measure to prevent PICD, but might prolong hospitalization in patients with large volumes of ascitic fluid.
The data of Coll et al.  reported in this issue of the Journal suggest a new approach to this problem. If confirmed in appropriate studies, reducing the flow rate of extraction of ascitic fluid could represent a low-cost, useful tool to decrease the incidence of PICD. Other measures, such as placing a pneumatic girdle during paracentesis, need further evaluation.
Plasma volume expanders have been shown to be effective in the prevention of PICD. As outlined above, the incidence of PICD reaches 80% when large-volume paracentesis is performed without additional therapeutic measures, and can be reduced to 15–35% when plasma volume expanders are used. The efficacy of each substance is related to its plasma half-life, those with a longer half-life being the most effective. The rationale for using plasma expanders after paracentesis is to refill the dilated vascular space after paracentesis, thereby preventing the subsequent activation of vasoconstrictor systems.
As outlined above, the incidence of PICD can be reduced to 15% when albumin is used as a plasma expander after paracentesis , and this is probably related to its prolonged half-life (21 days). However, albumin has disadvantages including its high cost and the suggestion that its use may be harmful in some cases . Other plasma expanders have been shown to be effective in the prevention of PICD: dextran 70 is a polysaccharide consisting of a mixture of glucose molecules with different molecular weight fractions ranging between 15 000 and 160 000 Da, with an elimination half-life of 24–48 h ; polygeline is a synthetic, polymerized gelatine with a half-life of 4 h; and saline solution is another short-life plasma expander that has been used after paracentesis. Several studies initially showed that these substances might represent a valid alternative to albumin, as the incidences of renal failure, complications and PRA variations encountered were similar to those experienced with albumin use, and they had the important advantage of being less costly [23,26–28]. However, a recent, large, randomized trial showed that albumin was better than dextran 70 and polygeline in the prevention of PICD . Furthermore, a recent randomized trial performed by our group  has shown that the incidence of PICD is significantly lower in patients receiving albumin than in those receiving saline. However, albumin was better than other plasma expanders only when a large volume of ascites was drained. In contrast, when a low volume of ascitic fluid was evacuated the incidence of PICD was low (about 10%) irrespective of the type of plasma expander used.
If paracentesis-induced arteriolar vasodilatation precedes and plays a major role in the development of decreased effective arterial blood volume, the administration of a splanchnic vasoconstrictor might be effective in the prevention of PICD. In a pilot study , 20 patients with tense ascites were randomly assigned to be treated with paracentesis plus terlipressin, a potent splanchnic vasoconstrictor, or paracentesis plus albumin. Terlipressin was administered as an intravenous 1 mg bolus at the onset of paracentesis and then 8 h and 16 h after the first bolus. No significant variations in plasma renin concentrations were observed irrespective of the treatment group and changes in baseline plasma renin concentrations did not differ between the terlipressin and albumin groups. However, because of the small sample size of the study group, a type II error cannot be excluded.
PICD is a clinically relevant complication of paracentesis. Despite initial studies suggesting that PICD was related to a decrease of total blood volume, there is now evidence that accentuation of arteriolar vasodilatation is the origin of this disorder. More studies designed to investigate PICD are needed. Plasma volume expansion is the only proven treatment to prevent the condition. More studies are needed to clarify the work of other therapeutic approaches, as detailed here, in the prevention of this disturbance.
Conflict of interest
1. Gines P, Arroyo V, Quintero E, Planas R, Bory F, Cabrera J, et al
. Comparison of paracentesis
and diuretics in the treatment of cirrhotics with tense ascites. Results of a randomized study. Gastroenterology
2. Quintero E, Gines P, Arroyo V, Rimola A, Bory F, Planas R, et al
versus diuretics in the treatment of cirrhotics with tense ascites. Lancet
3. Gines P, Tito L, Arroyo V, Planas R, Panes J, Viver J, et al
. Randomized comparative study of therapeutic paracentesis
with and without intravenous albumin in cirrhosis. Gastroenterology
4. Luca A, Garcia-Pagan JC, Bosch J, Feu F, Jimenez W, Gines A, et al
. Beneficial effects of intravenous albumin infusion on the hemodynamic and humoral changes after total paracentesis
5. Gines A, Fernandez-Esparrach G, Monescillo A, Vila C, Domenech E, Abecasis R, et al
. Randomized trial comparing albumin, dextran 70, and polygeline in cirrhotic patients with ascites treated by paracentesis
6. Knauer CM, Lowe HM. Hemodynamics in the cirrhotic patient during paracentesis
. N Engl J Med
7. Pozzi M, Osculati G, Boari G, Serboli P, Colombo P, Lambrughi C, et al
. Time course of circulatory and humoral effects of rapid total paracentesis
in cirrhotic patients with tense, refractory ascites. Gastroenterology
8. Simon DM, McCain JR, Bonkovsky HL, Wells JO, Hartle DK, Galambos JT. Effects of therapeutic paracentesis
on systemic and hepatic hemodynamics and on renal and hormonal function. Hepatology
9. Gentile S, Angelico M, Bologna E, Capocaccia L. Clinical, biochemical, and hormonal changes after a single, large-volume paracentesis
in cirrhosis with ascites. Am J Gastroenterol
10. Panos MZ, Moore K, Vlavianos P, Chambers JB, Anderson JV, Gimson AE, et al
. Single, total paracentesis
for tense ascites: sequential hemodynamic changes and right atrial size. Hepatology
11. Wang SS, Chen CC, Chao Y, Wu SL, Lee FY, Lin HC, et al
. Sequential hemodynamic changes for large volume paracentesis
in post-hepatitic cirrhotic patients with massive ascites. Proc Natl Sci Counc Repub China B
12. Pinto PC, Amerian J, Reynolds TB. Large-volume paracentesis
in nonedematous patients with tense ascites: its effect on intravascular volume. Hepatology
13. Salo J, Gines A, Gines P, Piera C, Jimenez W, Guevara M, et al
. Effect of therapeutic paracentesis
on plasma volume and transvascular escape rate of albumin in patients with cirrhosis. J Hepatol
14. Coll S, Vila MC, Molina L, Gimenez MD, Guarner C, Sola R. Mechanisms of early decrease in systemic vascular resistance after total paracentesis
: influence of flow rate of ascites extraction. Eur J Gastroenterol Hepatol
15. Cabrera J, Falcon L, Gorriz E, Pardo MD, Granados R, Quinones A, et al
. Abdominal decompression plays a major role in early postparacentesis haemodynamic changes in cirrhotic patients with tense ascites. Gut
16. Castro A, Jimenez W, Claria J, Ros J, Martinez JM, Bosch M, et al
. Impaired responsiveness to angiotensin II in experimental cirrhosis: role of nitric oxide. Hepatology
17. Ryan J, Sudhir K, Jennings G, Esler M, Dudley F. Impaired reactivity of the peripheral vasculature to pressor agents in alcoholic cirrhosis. Gastroenterology
18. Lunzer MR, Newman SP, Bernard AG, Manghani KK, Sherlock SP, Ginsburg J. Impaired cardiovascular responsiveness in liver disease. Lancet
19. Ruiz-del-Arbol L, Monescillo A, Jimenez W, Garcia-Plaza A, Arroyo V, Rodes J. Paracentesis-induced circulatory dysfunction
: mechanism and effect on hepatic hemodynamics in cirrhosis. Gastroenterology
20. Vila MC, Sola R, Molina L, Andreu M, Coll S, Gana J, et al
. Hemodynamic changes in patients developing effective hypovolemia after total paracentesis
. J Hepatol
21. Peltekian KM, Wong F, Liu PP, Logan AG, Sherman M, Blendis LM. Cardiovascular, renal, and neurohumoral responses to single large-volume paracentesis
in patients with cirrhosis and diuretic-resistant ascites. Am J Gastroenterol
22. Sola-Vera J, Minana J, Ricart E, Planella M, Gonzalez B, Torras X, et al
. Randomized trial comparing albumin and saline in the prevention of paracentesis-induced circulatory dysfunction
in cirrhotic patients with ascites. Hepatology
23. Planas R, Gines P, Arroyo V, Llach J, Panes J, Vargas V, et al
. Dextran-70 versus albumin as plasma expanders in cirrhotic patients with tense ascites treated with total paracentesis
. Results of a randomized study. Gastroenterology
24. Drummond GB, Ludlam CA. Is albumin harmful? Br J Haematol
25. Klotz U, Kroemer H. Clinical pharmacokinetic considerations in the use of plasma expanders. Clin Pharmacokinet
26. Fassio E, Terg R, Landeira G, Abecasis R, Salemne M, Podesta A, et al
with dextran 70 vs. paracentesis
with albumin in cirrhosis with tense ascites. Results of a randomized study. J Hepatol
27. Salerno F, Badalamenti S, Lorenzano E, Moser P, Incerti P. Randomized comparative study of Hemaccel vs. albumin infusion after total paracentesis
in cirrhotic patients with refractory ascites. Hepatology
28. Cabrera J, Inglada L, Quintero E, Jimenez W, Losada A, Mayor J, et al
. Large-volume paracentesis
and intravenous saline: effects on the renin–angiotensin system. Hepatology
29. Moreau R, Asselah T, Condat B, de Kerguenec C, Pessione F, Bernard B, et al
. Comparison of the effect of terlipressin and albumin on arterial blood volume in patients with cirrhosis and tense ascites treated by paracentesis
: a randomised pilot study. Gut