**7. Emerging value of determining FEurate**

Determination of serum urate in SIADH was first reported in 1971. Patients with SIADH had hypouricemia and increased FEurate, which normalized after correction of the hyponatremia. (Dorhout Mees et al, 1971) In 1979 Beck duplicated these findings, but compared serum urate in SIADH with other causes of hyponatremia. Except for only one overlapping value, there was complete separation of serum urate in SIADH from other causes of hyponatremia. (Beck, 1979) Correcting the hyponatremia by water restriction was accompanied by an increase in serum urate with normalization of a previously increased FEurate, figures 3, 4. (Beck, 1979) Beck concluded that the coexistence of hyponatremia and hypouricemia differentiated SIADH from most other causes of hyponatremia. This apparent simple method of differentiating SIADH from other causes of hyponatremia stimulated

Complexity of Differentiating Cerebral-Renal Salt Wasting

**SIADH RSW**

**BEFORE AFTER BEFORE AFTER CORRECTION OF HYPONATREMIA**

Figure 4 is updated version with elimination of SUA, see test.

from SIADH, Emerging Importance of Determining Fractional Urate Excretion 53

Fig. 3. Fig. 4. Fig. 3. and 4. Relationship between FEurate (FEUA ), serum urate (SUA ) and serum sodium (SNa ) before and after correction of hyponatremia in SIADH and RSW. Shaded areas represent normal ranges. (Modified from Maesaka, 1999), Figure 3.

We prospectively evaluated 96 patients with AIDS. Sixteen patients had combined hyponatremia and hypouricemia, 21 were hypouricemic, and 19 of 23 patients studied had increased FEurate. Many had increased FEurate with normonatremia, which was consistent with RSW and not SIADH. All hyponatremic patients had high UNa and concentrated urines that were consistent with SIADH and RSW. Ten hyponatremic patients with saline responsive postural hypotension, CVP of 0 cm H2O with high plasma renin and aldosterone

We extended our study to neurosurgical patients with diverse etiologies because 12 of 12 hypouricemia patients with AIDS had cortical atrophy of brain by CT scan. (Maesaka et al, 1990, 1992) The high incidence of RSW in neurosurgical patients suggested that RSW in AIDS might be due to their cerebral disease, so we prospectively studied urate metabolism in 29 neurosurgical patients of varying etiologies and 21 age and gender-matched controls. Seven patients were hypouricemic, 18 had FEurate > 10% and only 1 patient was hyponatremic. (Maesaka et al, 1992) We postulated that the increase in FEurate without hyponatremia was again consistent with RSW that was supported by the volume studies in neurosurgical patients. These observations solidified our notion that an increased FEurate with normonatremia may be consistent with RSW without a need to correct hyponatremia, figures 3, 4. Moreover, an increased FEurate was associated with serum urate > 5 mg/dL, suggesting a greater value of FEurate over serum urate. (Maesaka et al, 1992, 1998, 2009)

A study of urate metabolism focused on other cerebral diseases because patients with AD were reported to have lower serum urate. (Kasa et al, 1989) Serum urate was lower and FEurate higher in 18 patients with AD as compared to 6 patients with multi infarct dementia and 11 age and gender-matched normal controls. (Maesaka et al, 1993) As in the neurosurgical study, only one patient with AD was hyponatremic. The increased FEurate with normonatremia suggested that demented patients with AD might have RSW and raised the question whether or not an isolated increase in FEurate with normonatremia without prior correction of a pre-existing hyponatremia would be consistent with RSW.

collectively supported a diagnosis of RSW in AIDS. (Maesaka et al, 1990, 1990)

**BEFORE AFTER BEFORE AFTER CORRECTION OF HYPONATREMIA**

**SIADH RSW**

others to investigate this relationship and renal urate handling. Serum urate was consistently found to be decreased in SIADH and there was virtually no overlap or partial overlap with other causes of hyponatremia. (Assadi & John, 1985; Passamonte 1984; Sonnenblick et al, 1988; Sorensen et al, 1988). FEurate was similarly increased in SIADH. (Assadi, 1985; Beck, 1979; Bitew, 2009; Decaux, 1990; Dorhout Mees ,1971; Drakakis, 2011; Weinberger et al, 1982; Sonnenblick, 1988; Sorenson, 1988) Several studies also demonstrated normalization or reduction of a previously increased FEurate after correction of hyponatremia by water restriction, figure 3, 4. (Assadi & John, 1985; Beck, 1979; Decaux et al, 1990; DorhoutMees et al, 1971; Drakakis et al, 2011; Sonnenblick et al, 1988) Improvements in hypouricemia and increased FEurate after correction of hyponatremia, appear to be characteristic findings in SIADH, but the coexistence of hypouricemia and hyponatremia has not been found as useful as FEurate, figures 3, 4.

We encountered 5 patients, in whom an increased FEurate and hypouricemia persisted after correction of their hyponatremia by water-restriction, suggesting that these hyponatremic patients were pathophysiologically different from SIADH. (Maesaka et al, 1990) The first patient had metastatic pancreatic carcinoma with hypoalbuminemia, albumin 1.5 g/dL, edema, ascites, hypouricemia (serum urate 1.1 mg/dL), increased FEurate, 34.2%, hypophosphatemia, 1.7 mg/dL, with increased FEphosphate, 29.1%, UNa 99mmol/L and Uosm 716 mosm/kg. His FEurate remained persistently increased at 30.0% after correction of hyponatremia to 138 mmol/L by water-restriction. The edema, ascites and increased FEphosphate were inconsistent with SIADH and the collective findings were consistent with a variant of the Fanconi syndrome. The second case was a patient with bronchogenic carcinoma with a negative CT scan of brain, who presented with serum sodium of 116 mmol/L, saline-responsive postural hypotension, Uosm 323 mosm/kg, UNa 42 mmol/L, serum urate 2.0 mg/dL, FEurate 26.5% and normal renal, adrenal and thyroid function. A diagnosis of SIADH was made on the basis of the report by Beck and he was waterrestricted with liberal salt supplementation, which was followed by significant weight loss, recurrence of his postural hypotension, weakness, postural dizziness, slurred speech, staggered gait and somnolence. His serum sodium finally corrected to 138 mmol/l, and FEurate remained elevated at 14.7%, despite being severely volume depleted. At this time, his Uosm was 980 mosm/kg, UNa 181 mmol/L and he remained hypouricemic, serum urate 2.2 mg/dL. He responded well to saline therapy with reversal of all of his symptoms. The increased FEurate and hypouricemia persisted after correction of his hyponatremia by water restriction and salt supplementation, suggesting that a persistently elevated FEurate and hypouricemia after correction of hyponatremia would be consistent with RSW. We found a persistent increase in FEurate and hypouricemia after correction of hyponatremia in three additional cases, one with bronchogenic carcinoma that had metastasized to brain, another with disseminated Cryptococcus that involved brain and uncomplicated Hodgkins disease with no clinical cerebral disease. All had normal renal adrenal and thyroid function. The hyponatremia, high UNa and concentrated urine were consistent with SIADH except for the persistent increase in FEurate. The hyponatremia, high UNa and concentrated urine were consistent with SIADH except for the persistent increase in FEurate, which was construed as being pathophysiologically different from SIADH and might serve to differentiate SIADH from RSW, figure 3, 4. Moreover, the absence of clinical cerebral disease in 3 of the 5 patients raised questions regarding the appropriateness of the term cerebral salt wasting.

others to investigate this relationship and renal urate handling. Serum urate was consistently found to be decreased in SIADH and there was virtually no overlap or partial overlap with other causes of hyponatremia. (Assadi & John, 1985; Passamonte 1984; Sonnenblick et al, 1988; Sorensen et al, 1988). FEurate was similarly increased in SIADH. (Assadi, 1985; Beck, 1979; Bitew, 2009; Decaux, 1990; Dorhout Mees ,1971; Drakakis, 2011; Weinberger et al, 1982; Sonnenblick, 1988; Sorenson, 1988) Several studies also demonstrated normalization or reduction of a previously increased FEurate after correction of hyponatremia by water restriction, figure 3, 4. (Assadi & John, 1985; Beck, 1979; Decaux et al, 1990; DorhoutMees et al, 1971; Drakakis et al, 2011; Sonnenblick et al, 1988) Improvements in hypouricemia and increased FEurate after correction of hyponatremia, appear to be characteristic findings in SIADH, but the coexistence of hypouricemia and

We encountered 5 patients, in whom an increased FEurate and hypouricemia persisted after correction of their hyponatremia by water-restriction, suggesting that these hyponatremic patients were pathophysiologically different from SIADH. (Maesaka et al, 1990) The first patient had metastatic pancreatic carcinoma with hypoalbuminemia, albumin 1.5 g/dL, edema, ascites, hypouricemia (serum urate 1.1 mg/dL), increased FEurate, 34.2%, hypophosphatemia, 1.7 mg/dL, with increased FEphosphate, 29.1%, UNa 99mmol/L and Uosm 716 mosm/kg. His FEurate remained persistently increased at 30.0% after correction of hyponatremia to 138 mmol/L by water-restriction. The edema, ascites and increased FEphosphate were inconsistent with SIADH and the collective findings were consistent with a variant of the Fanconi syndrome. The second case was a patient with bronchogenic carcinoma with a negative CT scan of brain, who presented with serum sodium of 116 mmol/L, saline-responsive postural hypotension, Uosm 323 mosm/kg, UNa 42 mmol/L, serum urate 2.0 mg/dL, FEurate 26.5% and normal renal, adrenal and thyroid function. A diagnosis of SIADH was made on the basis of the report by Beck and he was waterrestricted with liberal salt supplementation, which was followed by significant weight loss, recurrence of his postural hypotension, weakness, postural dizziness, slurred speech, staggered gait and somnolence. His serum sodium finally corrected to 138 mmol/l, and FEurate remained elevated at 14.7%, despite being severely volume depleted. At this time, his Uosm was 980 mosm/kg, UNa 181 mmol/L and he remained hypouricemic, serum urate 2.2 mg/dL. He responded well to saline therapy with reversal of all of his symptoms. The increased FEurate and hypouricemia persisted after correction of his hyponatremia by water restriction and salt supplementation, suggesting that a persistently elevated FEurate and hypouricemia after correction of hyponatremia would be consistent with RSW. We found a persistent increase in FEurate and hypouricemia after correction of hyponatremia in three additional cases, one with bronchogenic carcinoma that had metastasized to brain, another with disseminated Cryptococcus that involved brain and uncomplicated Hodgkins disease with no clinical cerebral disease. All had normal renal adrenal and thyroid function. The hyponatremia, high UNa and concentrated urine were consistent with SIADH except for the persistent increase in FEurate. The hyponatremia, high UNa and concentrated urine were consistent with SIADH except for the persistent increase in FEurate, which was construed as being pathophysiologically different from SIADH and might serve to differentiate SIADH from RSW, figure 3, 4. Moreover, the absence of clinical cerebral disease in 3 of the 5 patients raised questions regarding the appropriateness of the term cerebral salt

hyponatremia has not been found as useful as FEurate, figures 3, 4.

wasting.

Fig. 3. and 4. Relationship between FEurate (FEUA ), serum urate (SUA ) and serum sodium (SNa ) before and after correction of hyponatremia in SIADH and RSW. Shaded areas represent normal ranges. (Modified from Maesaka, 1999), Figure 3. Figure 4 is updated version with elimination of SUA, see test.

We prospectively evaluated 96 patients with AIDS. Sixteen patients had combined hyponatremia and hypouricemia, 21 were hypouricemic, and 19 of 23 patients studied had increased FEurate. Many had increased FEurate with normonatremia, which was consistent with RSW and not SIADH. All hyponatremic patients had high UNa and concentrated urines that were consistent with SIADH and RSW. Ten hyponatremic patients with saline responsive postural hypotension, CVP of 0 cm H2O with high plasma renin and aldosterone collectively supported a diagnosis of RSW in AIDS. (Maesaka et al, 1990, 1990)

We extended our study to neurosurgical patients with diverse etiologies because 12 of 12 hypouricemia patients with AIDS had cortical atrophy of brain by CT scan. (Maesaka et al, 1990, 1992) The high incidence of RSW in neurosurgical patients suggested that RSW in AIDS might be due to their cerebral disease, so we prospectively studied urate metabolism in 29 neurosurgical patients of varying etiologies and 21 age and gender-matched controls. Seven patients were hypouricemic, 18 had FEurate > 10% and only 1 patient was hyponatremic. (Maesaka et al, 1992) We postulated that the increase in FEurate without hyponatremia was again consistent with RSW that was supported by the volume studies in neurosurgical patients. These observations solidified our notion that an increased FEurate with normonatremia may be consistent with RSW without a need to correct hyponatremia, figures 3, 4. Moreover, an increased FEurate was associated with serum urate > 5 mg/dL, suggesting a greater value of FEurate over serum urate. (Maesaka et al, 1992, 1998, 2009)

A study of urate metabolism focused on other cerebral diseases because patients with AD were reported to have lower serum urate. (Kasa et al, 1989) Serum urate was lower and FEurate higher in 18 patients with AD as compared to 6 patients with multi infarct dementia and 11 age and gender-matched normal controls. (Maesaka et al, 1993) As in the neurosurgical study, only one patient with AD was hyponatremic. The increased FEurate with normonatremia suggested that demented patients with AD might have RSW and raised the question whether or not an isolated increase in FEurate with normonatremia without prior correction of a pre-existing hyponatremia would be consistent with RSW.

Complexity of Differentiating Cerebral-Renal Salt Wasting

**7.2 Hyponatremia with normal FEurate in reset osmostat** 

FEurate increases in RSW.

from SIADH, Emerging Importance of Determining Fractional Urate Excretion 55

The increased FEurate in RSW, like that in SIADH, is presently not understood. There is a natriuretic factor(s) that is present in plasma and urine of patients with neurosurgical diseases and in plasma of patients with AD. As discussed above, this factor(s) has a major inhibitory effect on proximal tubule sodium transport. (Maesaka et al, 1993, 1993; Youmans & Maesaka, 2011) The proximal tubule is the exclusive site of urate reabsorption and predominant site for phosphate. It would be interesting to speculate that the natriuretic factor might have a dose-dependent effect with different affinities for various transporters in the proximal tubule. The circulating factor could also explain the persistent increase in FEurate after correction of hyponatremia in RSW. Consistent with this speculation is our patient with metastatic pancreatic cancer who had transport defects for sodium, urate and phosphate that were interpreted as being consistent with the Fanconi syndrome. (Maesaka et al, 1990) For the moment, however, there are few insights into mechanisms by which

Further appreciation of the emerging importance of determining FEurate in hyponatremic patients comes from our study of patients with RO or type C SIADH, which accounts for 36% of patients with SIADH. (Zerbe et al, 1980) It is characterized by euvolemia with normal renal, adrenal and thyroid function, hyponatremia resulting from ADH stimulation at a lower plasma osmolality or a RO, having a reasonably normal diluting and concentrating capacity of urine, and maintaining normal sodium balance without correcting the hyponatremia. (Wall, 1993) They are typically untreated, but this approach poses a therapeutic dilemma because of our tendency to treat most if not all hyponatremics. (DeFronzo et al, 1976; Decaux et al, 2009; Elisaf et al, 1990; Hill et al, 1990; Kahn, 2003) We encountered 3 patients with hyponatremia and normal FEurate, who were noted to excrete urines with Uosm < 200 mosm/kg, that was consistent with RO. Based on these findings, we decided to perform water-loading tests in nonedematous hyponatremic patients with normal FEurates without a dilute urine in a randomly collected urine, regardless of UNa or serum urate. (Imbriano, 2011) In this study of 14 patients, every nonedematous hyponatremic patient we encountered with a FEurate of 4-10% had RO as determined by Uosm <200 mosm/kg on a random urine collection, 8 patients, or after a normal waterloading test, 6 patients. As is typical of RO, plasma ADH was undetectable in 4 patients studied during the water-loading test. Eleven patients had UNa > 20 mmol/L, 8 were hypouricemic, yet all had a normal FEurate of < 10%. Interestingly, 3 patients were on losartan and 2 on atorvastatin, which increase urine urate excretion and lower serum urate. (Milionis et al, 2004; Yamamoto et al, 2000) These data suggest that chronic hyponatremia does not increase FEurate as has been proposed as a contributing factor for the increased FEurate in SIADH. (Decaux et al 2000) Comorbid conditions were similar to those reported in RO, including 2 patients without comorbidities. These and our other studies refine the proposal by Beck, that the coexistence of hyponatremia and hypouricemia differentiates SIADH from most other causes of hyponatremia, by stressing the greater value of FEurate over serum urate. (Beck, 1979; Maesaka et al, 1998, 2009) We concluded that RO occurs commonly, a normal FEurate in a nonedematous hyponatremic patient is highly suggestive of RO and FEurate has greater clinical utility than serum urate. (Imbriano et al, 2011) FEurate has a physiological basis for its derivation as compared to multiple factors that determine serum urate, including endogenous or exogenous purine sources, endogenous

One drawback to our proposal of determining FEurate after correction of hyponatremia to differentiate SIADH from RSW was the unpredictability of correcting the hyponatremia. We utilized the recommendation to treat hyponatremic patients with cerebral disease with hypertonic saline, regardless of whether or not the patient had SIADH or RSW. (Sterns et al, 2008) We increased serum sodium to 138 mmol/L within 2-3 days with 1.5% and determined FEurate, being aware that saline had only a modest effect on FEurate and would not likely alter our results. (Cannon, 1970; Diamond, 1975; Maesaka & Fishbane, 1998; Steele, 1974) We corrected hyponatremia within 2-3 days in 3 patients, who met usual criteria for SIADH, and found a previously increased FEurate to decrease to < 10% in all 3 patients. (Drakakis et al, 2011) Normalization of FEurate after correction of hyponatremia can be predictably achieved in 2-3 days by hypertonic saline and can be contrasted to a persistent increase in FEurate in RSW, figures 3, 4. The rate of correction can be controlled by monitoring serum sodium and adjusting free water intake to achieve the desired rate and extent of the correction of hyponatremia. These studies confirmed our notion that saline has a meager effect on FEurate by decreasing to normal levels during saline infusion. Under the proper circumstances of adequate cardiac function, hyponatremia can be corrected with 1.5% hypertonic saline in a predictably timely fashion to differentiate SIADH from RSW, figures 3, 4. (Drakakis et al, 2011)

#### **7.1 Pathophysiology of increased FEurate in SIADH and RSW**

The mechanisms by which FEurate increases in both SIADH and RSW are presently unexplained. The most prominent explanation has been the volume expansion that is seen in SIADH, but the saline infusion studies suggest that volume expansion has only a meager effect on FEurate. (Cannon et al, 1970; Diamond et al, 1975; Steele et al, 197) The V1 ADH receptor has been proposed as a cause for the increase in FEurate in SIADH after demonstrating an increase in FEurate by pitressin, and eliminating the effect by a V1-specific receptor inhibitor. (Decaux et al, 1996) This explanation is untenable because FEurate increased significantly at a time when normal subjects were hyponatremic during intranasal dDAVP, which lacks V1 activity, and by normalization of FEurate with correction of hyponatremia in SIADH when plasma ADH levels are still elevated. (Boer et al, 1987) It has been proposed that the defect in urate transport in SIADH was a result of an inhibition of the post secretory reabsorptive site for urate transport, based on the combination of the assumed secretory inhibition by pyrazinamide and decrease in post secretory reabsorption of urate by sulfapyrazone. (Weinberger et al, 1982) Studies in brush border membranes, however, demonstrate increased uptake of urate in the presence of pyrazinamide, suggesting that the decrease in urate excretion during pyrazinamide administration was a result of increased reabsorption and not inhibition of secretion. (Roch-Ramel et al, 1994) Until there is further clarification of the mechanisms and sites at which urate is transported, we must conclude that there is increased FEurate at a time when the patient is hyponatremic in SIADH, but the mechanism for this unique transport abnormality remains unclear. Lastly, the proposal that the chronic hyponatremia or hypo-osmolality contributed to the increase in FEurate in SIADH cannot be supported by the normal FEurate that has been reported in hyponatremia due to psychogenic polydipsia and more recently, RO, to be discussed below. (Ali et al, 2009; Decaux et al, 2000; Imbriano et al, 2011)

One drawback to our proposal of determining FEurate after correction of hyponatremia to differentiate SIADH from RSW was the unpredictability of correcting the hyponatremia. We utilized the recommendation to treat hyponatremic patients with cerebral disease with hypertonic saline, regardless of whether or not the patient had SIADH or RSW. (Sterns et al, 2008) We increased serum sodium to 138 mmol/L within 2-3 days with 1.5% and determined FEurate, being aware that saline had only a modest effect on FEurate and would not likely alter our results. (Cannon, 1970; Diamond, 1975; Maesaka & Fishbane, 1998; Steele, 1974) We corrected hyponatremia within 2-3 days in 3 patients, who met usual criteria for SIADH, and found a previously increased FEurate to decrease to < 10% in all 3 patients. (Drakakis et al, 2011) Normalization of FEurate after correction of hyponatremia can be predictably achieved in 2-3 days by hypertonic saline and can be contrasted to a persistent increase in FEurate in RSW, figures 3, 4. The rate of correction can be controlled by monitoring serum sodium and adjusting free water intake to achieve the desired rate and extent of the correction of hyponatremia. These studies confirmed our notion that saline has a meager effect on FEurate by decreasing to normal levels during saline infusion. Under the proper circumstances of adequate cardiac function, hyponatremia can be corrected with 1.5% hypertonic saline in a predictably timely fashion to differentiate SIADH from RSW,

The mechanisms by which FEurate increases in both SIADH and RSW are presently unexplained. The most prominent explanation has been the volume expansion that is seen in SIADH, but the saline infusion studies suggest that volume expansion has only a meager effect on FEurate. (Cannon et al, 1970; Diamond et al, 1975; Steele et al, 197) The V1 ADH receptor has been proposed as a cause for the increase in FEurate in SIADH after demonstrating an increase in FEurate by pitressin, and eliminating the effect by a V1-specific receptor inhibitor. (Decaux et al, 1996) This explanation is untenable because FEurate increased significantly at a time when normal subjects were hyponatremic during intranasal dDAVP, which lacks V1 activity, and by normalization of FEurate with correction of hyponatremia in SIADH when plasma ADH levels are still elevated. (Boer et al, 1987) It has been proposed that the defect in urate transport in SIADH was a result of an inhibition of the post secretory reabsorptive site for urate transport, based on the combination of the assumed secretory inhibition by pyrazinamide and decrease in post secretory reabsorption of urate by sulfapyrazone. (Weinberger et al, 1982) Studies in brush border membranes, however, demonstrate increased uptake of urate in the presence of pyrazinamide, suggesting that the decrease in urate excretion during pyrazinamide administration was a result of increased reabsorption and not inhibition of secretion. (Roch-Ramel et al, 1994) Until there is further clarification of the mechanisms and sites at which urate is transported, we must conclude that there is increased FEurate at a time when the patient is hyponatremic in SIADH, but the mechanism for this unique transport abnormality remains unclear. Lastly, the proposal that the chronic hyponatremia or hypo-osmolality contributed to the increase in FEurate in SIADH cannot be supported by the normal FEurate that has been reported in hyponatremia due to psychogenic polydipsia and more recently, RO, to be discussed below.

figures 3, 4. (Drakakis et al, 2011)

**7.1 Pathophysiology of increased FEurate in SIADH and RSW** 

(Ali et al, 2009; Decaux et al, 2000; Imbriano et al, 2011)

The increased FEurate in RSW, like that in SIADH, is presently not understood. There is a natriuretic factor(s) that is present in plasma and urine of patients with neurosurgical diseases and in plasma of patients with AD. As discussed above, this factor(s) has a major inhibitory effect on proximal tubule sodium transport. (Maesaka et al, 1993, 1993; Youmans & Maesaka, 2011) The proximal tubule is the exclusive site of urate reabsorption and predominant site for phosphate. It would be interesting to speculate that the natriuretic factor might have a dose-dependent effect with different affinities for various transporters in the proximal tubule. The circulating factor could also explain the persistent increase in FEurate after correction of hyponatremia in RSW. Consistent with this speculation is our patient with metastatic pancreatic cancer who had transport defects for sodium, urate and phosphate that were interpreted as being consistent with the Fanconi syndrome. (Maesaka et al, 1990) For the moment, however, there are few insights into mechanisms by which FEurate increases in RSW.

### **7.2 Hyponatremia with normal FEurate in reset osmostat**

Further appreciation of the emerging importance of determining FEurate in hyponatremic patients comes from our study of patients with RO or type C SIADH, which accounts for 36% of patients with SIADH. (Zerbe et al, 1980) It is characterized by euvolemia with normal renal, adrenal and thyroid function, hyponatremia resulting from ADH stimulation at a lower plasma osmolality or a RO, having a reasonably normal diluting and concentrating capacity of urine, and maintaining normal sodium balance without correcting the hyponatremia. (Wall, 1993) They are typically untreated, but this approach poses a therapeutic dilemma because of our tendency to treat most if not all hyponatremics. (DeFronzo et al, 1976; Decaux et al, 2009; Elisaf et al, 1990; Hill et al, 1990; Kahn, 2003) We encountered 3 patients with hyponatremia and normal FEurate, who were noted to excrete urines with Uosm < 200 mosm/kg, that was consistent with RO. Based on these findings, we decided to perform water-loading tests in nonedematous hyponatremic patients with normal FEurates without a dilute urine in a randomly collected urine, regardless of UNa or serum urate. (Imbriano, 2011) In this study of 14 patients, every nonedematous hyponatremic patient we encountered with a FEurate of 4-10% had RO as determined by Uosm <200 mosm/kg on a random urine collection, 8 patients, or after a normal waterloading test, 6 patients. As is typical of RO, plasma ADH was undetectable in 4 patients studied during the water-loading test. Eleven patients had UNa > 20 mmol/L, 8 were hypouricemic, yet all had a normal FEurate of < 10%. Interestingly, 3 patients were on losartan and 2 on atorvastatin, which increase urine urate excretion and lower serum urate. (Milionis et al, 2004; Yamamoto et al, 2000) These data suggest that chronic hyponatremia does not increase FEurate as has been proposed as a contributing factor for the increased FEurate in SIADH. (Decaux et al 2000) Comorbid conditions were similar to those reported in RO, including 2 patients without comorbidities. These and our other studies refine the proposal by Beck, that the coexistence of hyponatremia and hypouricemia differentiates SIADH from most other causes of hyponatremia, by stressing the greater value of FEurate over serum urate. (Beck, 1979; Maesaka et al, 1998, 2009) We concluded that RO occurs commonly, a normal FEurate in a nonedematous hyponatremic patient is highly suggestive of RO and FEurate has greater clinical utility than serum urate. (Imbriano et al, 2011) FEurate has a physiological basis for its derivation as compared to multiple factors that determine serum urate, including endogenous or exogenous purine sources, endogenous

Complexity of Differentiating Cerebral-Renal Salt Wasting

on first encounter with the patient.

from SIADH, Emerging Importance of Determining Fractional Urate Excretion 57

2002)Even under ideal circumstances the delay in obtaining these results does not assist us

It is evident that the evaluation of the volume status or determining UNa has limited utility. This dilemma persists to this day on first encounter with the nonedematous hyponatremic patient. We developed an algorithm, which emphasizes FEurate in the evaluation of hyponatremia, table 1. We propose with supporting data to determine FEurate in any nonedematous patient with hyponatremia. If the FEurate is 5-10%, we should consider psychogenic polydipsia, RO and prerenal causes, such as congestive heart failure, cirrhosis, nephrosis and pre eclampsia, or hypovolemia with normal renal function. Psychogenic polydipsia can be eliminated from the history of increased water intake and very dilute urines. (Ali et al, 2009) Edematous states can be ruled out by the presence of edema. The major obstacle might be the hypovolemic patient with normal renal function and classic prerenal azotemia, but the low mean FEurate of 2.85%, possibly high serum urate and UNa < 20 mmol/l might assist in differentiating this group from RO. (Steele, 1969) In patients who do not fall into the groups mentioned, we would consider RO and search diligently for a dilute urine in a random urine collection. In the absence of a dilute random urine, we do not recommend performing a water-loading test to prove the diagnosis of RO. (Imbriano et al, 2011) We would instead treat them with water restriction and salt supplementation, however unsuccessful they are, and consider using ADH receptor inhibitors, vaptans.

If the FEurate exceeds preferably 12%, there are three possibilities to consider, SIADH, RSW, thiazide diuretics and drugs that induce an SIADH-like picture. Thiazide diuretics and neurotropics can be readily eliminated by a proper history so the major differential would be SIADH and RSW. We propose to correct serum sodium either by water-restriction with salt supplementation or 1.5% hypertonic saline and determine FEurate. (Drakakis et al, 2011) If FEurate corrects to < 10%, we would proceed with treatment for SIADH or if it exceeds 10%, preferably >12%, we would treat the patient for RSW with saline. The question represented as dotted lines in table 1 depends on whether the coexistence of increased FEurate with normonatremia would be diagnostic of RSW. While there are supporting data to suggest this to be a valid conclusion, especially in patients with neurosurgical diseases, future studies will hopefully provide further insights into this relationship. Because neurosurgical patients are routinely treated with hypertonic saline, an increased FEurate with normonatremia or hypernatremia would be suggestive of RSW, see above. We have found this algorithm to be superior to the previous approach as discussed above and expect

to make further refinements to this algorithm in the future.

**10. Treatment of the hyponatremic patient with SIADH and RSW** 

The increasing reports of significant symptoms being attributed to even mild hyponatremia not only shed light on a long unrecognized phenomenon, but introduce the need for some urgency in developing adequate treatment strategies for a condition with diverse etiologies and divergent therapeutic goals. Treatment, however, has undergone a period of uncertainty due to adverse outcomes that are related to delays in treatment and overcorrection of chronic hyponatremia. (Berl et al, 1990) The approach to methods of correction in different clinical conditions will be limited to SIADH and RSW and the reader is referred to broader reviews of treating hyponatremia. (Stern et al, 2009; Verbalis et al, 2007) As discussed earlier, foremost among the diagnostic and therapeutic dilemma is the need to

production and excretion via gut and urine. Moreover, the arbitrary definition of hypouricemia ranging from 1.5-4 mg/dL reflect the uncertainty of the value of serum urate as compared to well-defined parameters that have been established for FEurate. (Beck, 1979; Maesaka et al, 1990 e; Ramsdell & Kelley, 1973) These studies also introduce the possibility that RO is pathophysiologically different from the more traditional SIADH by virtue of the normal FEurate and predictability of ADH responses to plasma osmolality.
