Abstract and Introduction
Abstract
Context: Differential diagnosis of thiazide-associated hyponatremia (TAH) is challenging. Patients can either have volume depletion or a syndrome of inappropriate antidiuresis (SIAD)-like presentation.
Objective: To evaluate the impact of the simplified apparent strong ion difference in serum (aSID; sodium + potassium − chloride) as well as the urine chloride and potassium score (ChU; chloride − potassium in urine) in the differential diagnosis of TAH, in addition to assessment of fractional uric acid excretion (FUA).
Methods: Post hoc analysis of prospectively collected data from June 2011 to August 2013 from 98 hospitalized patients with TAH < 125 mmol/L enrolled at University Hospital Basel and University Medical Clinic Aarau, Switzerland. Patients were categorized according to treatment response in volume-depleted TAH requiring volume substitution or SIAD-like TAH requiring fluid restriction. We computed sensitivity analyses with ROC curves for positive predictive value (PPV) and negative predictive value (NPV) of aSID, ChU, and FUA in differential diagnosis of TAH.
Results: An aSID > 42 mmol/L had a PPV of 79.1% in identifying patients with volume-depleted TAH, whereas a value < 39 mmol/L excluded it with a NPV of 76.5%. In patients for whom aSID was inconclusive, a ChU < 15 mmol/L had a PPV of 100% and a NPV of 83.3%, whereas FUA < 12% had a PPV of 85.7% and a NPV of 64.3% in identifying patients with volume-depleted TAH.
Conclusion: In patients with TAH, assessment of aSID, potassium, and chloride in urine can help identifying patients with volume-depleted TAH requiring fluid substitution vs patients with SIAD-like TAH requiring fluid restriction.
Introduction
Patients undergoing thiazide or thiazide-like treatment are at risk for hyponatremia,[1,2] with a reported incidence of hyponatremia up to 30% in hospitalized patients.[3] Patients with thiazide- or thiazide-like-associated hyponatremia (TAH) can either present with a diuretic-induced volume depletion and consequent hypovolemic hyponatremia, or with a picture of euvolemic hyponatremia mimicking a syndrome of inappropriate antidiuresis (SIAD), in which case it is referred to as SIAD-like TAH.[4,5]
The European Clinical Practice Guidelines[6] for diagnosis and management of hypotonic hyponatremia propose a diagnostic algorithm based on urine indices (especially the measurement of urine osmolality and sodium) instead of the classical volume status assessment, of which the low sensitivity and specificity are well recognized.[7] However, the use of urine indices is of little utility in patients with TAH, because thiazide and thiazide-like diuretics promote natriuresis and alter urine dilution.[8] Fenske et al proposed the use of fractional uric acid excretion (FUA) in the differential diagnosis of TAH;[9] however, their recommendation was based on data from only 7 patients.
Thiazide and thiazide-like diuretics exert their mechanism of action not only on sodium, but also on chloride and potassium. We therefore aimed to evaluate the possible role of the imbalance between these 3 ions in the differential diagnosis of TAH. To do so, we calculated the simplified strong ion difference (serum sodium + potassium − chloride), a part of the Stewart model for acid-base disturbances.[10,11] In brief, the Stewart model represents an alternative to bicarbonate-centered approaches and is especially used in anesthesiology and intensive care. According to Stewart, acid-base balance is the result of the strong ions difference (SID, calculated as the sum of measured cations minus the sum of measured anions), weak acids, and partial pressure of carbon dioxide (pCO2), and aims to maintain electroneutrality in extracellular fluid.[12] In stable patients with no critical illness (absence of severe hypoalbuminemia, of acute renal failure and/or respiratory failure), the simplified aSID could be sufficient to evaluate acid-base balance. A value >40 mmol/L identifies patients with fluid depletion and contraction alkalosis due to relative hypochloremia, whereas a value <40 mmol/L identifies patients with acidosis due to relative excess of water and therefore in need of fluid restriction.[12]
Chloride and potassium balance in urine represents a sodium-independent marker of kidney response to diuretic-induced volume depletion.[13] Processes going on at the level of intercalated cells are not completely understood, but there is increasing evidence of a sodium-independent chloride reabsorption in response to changes in volume and metabolic status.[14,15] A low urine chloride is a known sign of hypovolemia.[16] It triggers renin release and intercalated cells activation, thus aggravating potassium depletion caused by thiazide.[17] Therefore, we further aimed to investigate the diagnostic potential of the balance between chloride and potassium in urine in the differential diagnosis of TAH. The goal of our analysis was to find an additional tool for the initial assessment of profound hyponatremia in the context of hospitalization and unreliable canonical urine indices.
J Clin Endocrinol Metab. 2023;108(9):2248-2254. © 2023 Endocrine Society
