The portocardiorenal axis and refractory ascites: The underfilled cup runneth over

Plasma immunoreactive alpha-human atrial natriuretic peptide (ANP) was measured in six cirrhotic patients with massive refractory ascites, under strict metabolic conditions, while they were receiving a 20-meq sodium diet, both before and at two-hour intervals for eight hours following peritoneovenous shunting (PVS). The mean preoperative level of ANP was 75±18pg/ml, which was found to be significantly higher than the normal range for this laboratory *8 to 24pg/ml) (p<0.05). This value was also significantly higher than the value of 21±5pg/ml (p<0.05) obtained in six patients with cirrhosis but without ascites. Following shunt insertion, an immediate natriuresis and diuresis were observed in five of the six cirrhotic patients with refractory ascites. In these five, right atrial pressure and ANP rose immediately, followed by a rise in the level of urinary cyclic guanosine monophosphate. The sixth subject had a delayed rise in right atrial pressure, and correspondingly the rise in ANP, the diuresis, and natriuresis were with changes in right atrial pressure (p<0.05), urinary cyclic guanosine monophosphate (p<0.05), urinary sodium excretion (p<0.05), and urine volume (p<0.05). These results suggest that ANP may be important in mediating the acute response to PVS. Human atrial natriuretic factor (ANF) levels were measured before and after peritoneovenous shunt implantation in 10 cirrhotic patients with ascites in whom sodium retention is a major clinical problem. The mean preoperative plasma level of ANF was 82ng/L (normal range, 5-80ng/L). Peritoneovenous shunting resulted in a significant rise in plasma ANF to 308ng/L (<0.0025) immediately after operation. This was followed by a constant fall until the seventh postoperative day, when mean plasma ANF was still significantly elevated (149ng/L) compared with the preoperative value (p<0.01). Three months after shunt implantation mean plasma ANF had returned to the preoperative level (75ng/L). Mean sodium excretion increased from 2.6mEq/h peoperatively to 10.2mEq/h at the second postoperative day (p<0.025). No direct relationship was noted between changes in plasma ANF level and changes in urinary sodium excretion after shunt implantation. These data demonstrate that intact ANF release response to intravascular volume expansion in cirrhotic patients with ascites, but exclude ANF as the diminished natriuretic factor as proposed by the overflow theory of ascites formation. Sodium excretion and fluid retention seem to be the result of vascular underfill and fluid maldistribution, and hormonal changes are likely to be secondary to them.