Still Learning from Our Patients: Hypokalemia in Patients with Lupus Nephritis

Parkland Memorial Hospital is a county hospital in Dallas, Texas, that provides care to a medically underserved population. Lupus nephritis is a common diagnosis, as I learned during my time there as a renal fellow and junior faculty member. In 2012, Dr. Kamalanathan Sambandam established a GN clinic at Parkland. Since inception, the clinic has cared for 435 patients with lupus nephritis, and an additional 464 patients with other glomerular diseases (K. Sambandam, personal communication). In this issue of Kidney360, Adomako and colleagues (¹) draw on their clinical experience caring for this large population of patients to describe a new entity, idiopathic hypokalemia in lupus nephritis. They identify 20 patients with this disorder, who do not have evidence of other suspected or known causes of hypokalemia, for a prevalence of 5%. Renal potassium wasting is demonstrated in a subset of patients by measurement of the urine potassium/creatinine ratio, transtubular potassium gradient, or measurement of 24-hour urinary potassium excretion.

What is the etiology of hypokalemia in these patients? Hypokalemia can occur in renal tubular acidosis (RTA), which has been previously described in patients with lupus nephritis. The authors therefore study ten patients from their clinic with lupus nephritis and RTA (prevalence, 3%). Not surprisingly, compared with control patients with lupus nephritis but without evidence of tubulopathy, the patients with RTA had a lower serum bicarbonate and higher urine pH, and four of ten patients had evidence of nephrocalcinosis or nephrolitihiasis. Although serum potassium was slightly lower in patients with RTA (3.75 mmol/L) compared with controls (4.00 mmol/L), this difference did not reach statistical significance. In contrast, the patients with idiopathic hypokalemia had lower serum potassium (3.26 mmol/L), normal serum bicarbonate and urine pH, and none had nephrocalcinosis or nephrolitihiasis. As described below, the patients with hypokalemia also had a distinct profile of serological markers. Although the diagnosis of incomplete distal RTA was not ruled out in the patients who were hypokalemic using provocative testing, the tubulopathy in these patients appears to be distinct from RTA.

The authors exclude other possible causes of hypokalemia by history and laboratory evaluation. Chronic diarrhea was excluded by history, and, as noted above, many of the hypokalemic patients had demonstrated renal potassium wasting. Primary hyperaldosteronism was excluded in a subset of patients by measurement of renin activity and aldosterone concentrations. In most patients, plasma aldosterone was below the limit of detection or low, as expected given hypokalemia. Corticosteroid exposure was lower in the patients who were hypokalemic compared with controls with lupus nephritis, so mineralocorticoid effects of steroids are unlikely. Data were censored when patients were exposed to diuretics or alkali therapy; thus, these were not contributors. Data were also censored for patients with conditions or treatments that can increase serum potassium, including eGFR <60 ml/min per 1.73 m² or during periods of acute kidney injury, and when patients were exposed to potassium supplements or calcineurin inhibitors. Because most patients were on angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, including 85% of the patients with hypokalemia, data were not censored on the basis of use of these medications, and, in any case, hypokalemia persisted despite their use. In fact, all patients with hypokalemia required either potassium supplementation or mineralocorticoid antagonist therapy to maintain normal serum potassium concentrations.

In the aldosterone-sensitive distal nephron, sodium reabsorption through the epithelial sodium channel generates the lumen-negative charge that contributes to a favorable electrochemical gradient for potassium secretion in the connecting tubule and collecting duct. Nephrotic proteinuria has been proposed to increase activation of the epithelial sodium channel, due to increased presence of urinary proteases that can activate the channel by cleavage (²). However, the patients described by Adomako et al. with hypokalemia had, on average, subnephrotic proteinuria (0.73 g/g protein/creatinine) that was similar to controls, so this is unlikely to be the etiology of hypokalemia in these patients.

A greater proportion of the patients with lupus nephritis with hypokalemia are described as Black (50%, compared with 20% in controls). Excess activity of the epithelial sodium channel has been suggested to underlie increased salt-sensitive hypertension in individuals of African descent (³), and could also contribute to hypokalemia. Consistent with this idea, one of five patients in whom plasma renin activity and aldosterone concentration were measured had suppressed renin activity (<0.6 ng/m per hr) and undetectable aldosterone (<1.6 ng/ml), which could be suggestive of elevated activity of the epithelial sodium channel. It would be interesting to determine whether this patient’s blood pressure was elevated, and whether amiloride, an inhibitor of the epithelial sodium channel, would be beneficial in the treatment of hypokalemia in this patient.

A striking observation in this study is the distinct autoimmune profiles in the three groups of patients studied. Those with hypokalemia had increased anti-Ro/SSA and anti-RNP antibodies compared with controls, whereas patients with RTA had increased anti-Ro/SSA and anti-La/SSB antibodies. This may reflect racial differences in autoantibody profiles, as discussed by the authors, but is also intriguing given case reports describing acquired Gitelman syndrome (^456–7) and isolated hypokalemia (⁸) in patients with Sjögren’s syndrome. Consistent with the idea of acquired Gitelman syndrome, the patients who were hypokalemic described by Adomako et al. had a mild, but statistically significant, decrease in serum magnesium compared with controls (1.73 mg/dl compared with 2.00 mg/dl). This decrease in serum magnesium is not explained by differences in use of proton pump inhibitors or the presence of diarrhea. Indeed, several of the patients from the reports of acquired Gitelman syndrome also had normal or only modestly decreased serum magnesium. Also consistent with a possible diagnosis of acquired Gitelman syndrome, two of the patients reported by Adomako et al. had elevated renin activity, which could be consistent with a salt-losing tubulopathy. Determination of other indicators of Gitelman syndrome, such as elevated fractional excretion of magnesium and chloride and hypocalciuria (⁹), would be of interest. The demonstration of decreased or absent sodium-chloride cotransporter in the distal convoluted tubule, and/or the presence of antibodies against the sodium-chloride cotransporter, as has been previously demonstrated in a patient with Sjögren syndrome and acquired Gitelman syndrome (⁴), would provide further evidence, and may be possible given that all patients described have previously undergone kidney biopsy.

The finding that the patients with lupus nephritis-associated RTA in this report were more likely to have increased anti-Ro/SSA and anti-La/SSB antibodies also suggests a possible connection to Sjögren’s syndrome, in which distal RTA is a well-described occurrence. The authors note that overlap syndromes, defined as lupus concomitant with other rheumatological disorders, including Sjögren’s syndrome, were more common in the patients with RTA, but did not reach statistical significance; however, this could be due to the small number of patients with RTA studied. Ongoing study of a larger group of patients could help determine whether distal RTA is more likely in patients with overlap syndrome, especially Sjögren’s, or whether there are similar patterns of autoantibodies that can arise that affect tubular function in the two disorders, even if clinical manifestations of both are not apparent. Whether there is a correlation between tubulointerstitial disease on renal biopsy and either hypokalemia or RTA would also be interesting to examine. Another question is whether idiopathic hypokalemia can occur in patients who have systemic lupus erythematosus without nephritis.

In summary, Adomako et al. describe a novel renal tubulopathy resulting in hypokalemia in patients with lupus nephritis, an entity first described in 1902 (¹⁰). Although the pathogenesis of this new electrolyte disorder has not been entirely elucidated, and may differ between patients, there are intriguing clues that may inform future studies in this area. Importantly, the information presented was all obtained during routine clinical practice in a large clinic dedicated to the care of patients with lupus nephritis and other glomerular disorders. Perhaps one of the most important lessons of this interesting paper is what clinicians can learn from careful study of the patients they see every day.

Disclosures

A. Rodan reports receiving honoraria from Augusta University, Charité Hospital, National Defense Medical Center (Taiwan), National Institute for Physiological Sciences (Japan), National Taiwan University, Tri-Service General Hospital (Taiwan), University of Alabama Birmingham, Universität Osnabrück, and University of Texas Southwestern, for visiting professorships at academic institutions; reports being a scientific advisor or member of American Journal of Physiology Renal Physiology Editorial Board and Kidney360 Editorial Board; and reports other interests/relationships with ASN Nominating Committee, National Institutes of Health grant reviewer, and National Science Foundation grant reviewer.

Funding

This work is supported by the HHS, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, R01 DK110358