A 42-Year-Old Woman With Flaccid Paralysis

Abstract

Editors' Note: This is the second in a series of invited Acid-Base and Electrolyte Teaching Cases. Advisory Board member Glenn Nagami, MD, served as the Consulting Editor for this case. Editors' Note: This is the second in a series of invited Acid-Base and Electrolyte Teaching Cases. Advisory Board member Glenn Nagami, MD, served as the Consulting Editor for this case. Normal-anion-gap (hyperchloremic) metabolic acidosis occurs commonly in the clinical setting. The severity of the acidosis may vary, but characteristic features include a normal serum anion gap, hypobicarbonatemia, and acidemia (depending on the degree of respiratory compensation). The approach necessitates stepwise analysis for accurate diagnosis. This case emphasizes that the segregation of these disorders into those of renal and nonrenal origin requires appreciation of the response by the kidney to the prevailing acidosis and to discern whether the patient's response is appropriate or inappropriate. The use of urinary electrolyte levels as a surrogate means of predicting net acid excretion is described to assist in the categorization of these disorders. A 42-year-old Native American woman presented to a local hospital with flaccid paralysis and severe hypokalemia (potassium level, 2.0 mEq/L [2.0 mmoI/L]). An intravenous infusion containing potassium chloride (20 mEq/L) was initiated, and the patient was transferred to an academic medical center. The patient's laboratory data are listed in Table 1.Table 1Laboratory DataParameterValueSodium (mEq/L)140Potassium (mEq/L)2.5Chloride (mEq/L)125Bicarbonate (mEq/L)5Anion gap (mEq/L)10Serum urea nitrogen (mg/dL)28Serum creatinine (mg/dL)1.7Estimated glomerular filtration rate (mL/min/1.73 m2)48Arterial blood gas pH7.11 Paco216 Bicarbonate (mEq/L)5 Pao2 (mm Hg)90 Calcium (mg/dL)8.5 Magnesium (mEq/L)2.5Note: Conversion factors for units: serum urea nitrogen in mg/dL to mmol/L, ×0.357; serum creatinine in mg/dL to μmol/L, ×88.4; estimated glomerular filtration rate in mL/min/1.73 m2 to mL/s/1.73 m2, ×0.01667; calcium in mg/dL to mmol/L, ×0.2495; magnesium in mEq/L to mmol/L, ×0.5. No conversion necessary for serum sodium, potassium, chloride, bicarbonate, and anion gap expressed in mEq/L and mmol/L. Open table in a new tab Note: Conversion factors for units: serum urea nitrogen in mg/dL to mmol/L, ×0.357; serum creatinine in mg/dL to μmol/L, ×88.4; estimated glomerular filtration rate in mL/min/1.73 m2 to mL/s/1.73 m2, ×0.01667; calcium in mg/dL to mmol/L, ×0.2495; magnesium in mEq/L to mmol/L, ×0.5. No conversion necessary for serum sodium, potassium, chloride, bicarbonate, and anion gap expressed in mEq/L and mmol/L. The patient was seen in consultation by the nephrology service, and urinalysis showed pH of 6.0, trace protein, and sediment with 5 to 10 white blood cells and no red blood cells per high power field and no bacteria or casts. A more focused interview by the attending physician uncovered a 3- to 4-year history of zerostomia and keratoconjunctivitis sicca, but no history of synovitis, arthritis, or rash. Urine protein-to-creatinine ratio was 0.150 g/g. Urinary electrolyte levels were as follows: sodium, 35 mEq/L (35 mmol/L); potassium, 40 mEq/L (40 mmol/L); chloride, 18 mEq/L (18 mmol/L; urine anion gap [UAG], +57 mEq/L [+57 mmol/L]); and urinary ammonium, 2 mEq/L. Schirmer test result was positive (indicating insufficient tear production), and serological data showed positive results for rheumatoid factor, anti-Ro/SS-A, and anti-La/SS-B. Test results for hepatitis B surface antigen and hepatitis C antibodies were negative. The kidneys measured 10 cm in length using ultrasound, with slightly increased echogenicity, but no evidence of obstruction. No areas of calcification were identified. Estimated glomerular filtration rate did not improve with hydration, and a kidney biopsy was performed. As shown in Fig 1 (light microscopy), there was a prominent interstitial infiltrate coalescing around collecting tubules and consisting predominately of lymphocytes. Primary Sjögren syndrome with classic distal renal tubular acidosis (RTA) and flaccid paralysis caused by hypokalemia. After replacement of the potassium deficit with intravenous potassium chloride in 0.45% sodium chloride (total, 60 mEq over 18 hours) and correction of the normal-anion-gap (hyperchloremic) metabolic acidosis with intravenous sodium bicarbonate, the patient was started on oral alkali replacement therapy consisting of Shohl's solution (sodium citrate solution [500 mg] and citric acid [334 mg]), 1 tablespoon 3 times daily. The flaccid paralysis resolved with potassium repletion, and there were no recurrent episodes of paralysis. Because of the sicca complex and intense lymphocytic infiltrate on kidney biopsy (Fig 1), the patient was followed up by the rheumatology division and started on prednisone and cyclophosphamide treatment, but was lost to follow-up by the nephrology service. The last available electrolyte levels were as follows: sodium, 140 mEq/L (140 mmol/L); potassium, 4.0 mEq/L (4.0 mmol/L); chloride, 106 mEq/L (106 mmol/L); and bicarbonate, 23 mEq/L (23 mmol/L). Potassium supplementation usually is unnecessary in adult patients on chronic alkali replacement therapy because re-expansion of extracellular fluid inhibits the elaboration of renin-angiotensin-aldosterone, typical of volume depletion in untreated patients with distal RTA. The first step in the evaluation is to appreciate the distinction between nonrenal and renal origins of normal-anion-gap (hyperchloremic) metabolic acidoses. This patient showed many diagnostic features of classic distal RTA (Box 1). The kidney is responsible for reabsorption of the filtered bicarbonate load (∼4,000 mEq/d) and production of new bicarbonate through net acid excretion (∼70 mEq/d). Both processes require hydrogen ion secretion by proximal and distal nephron segments. Net acid excretion is highly responsive to systemic acidosis, and upregulation of ammonium production and excretion is expected in patients with all nonrenal forms of normal-anion-gap (hyperchloremic) metabolic acidosis. In the case presented here, urinary ammonium excretion was not upregulated, evidenced by a positive UAG and low urine ammonium concentration. The level of urinary ammonium excretion in metabolic acidosis can be assessed indirectly by calculating UAG1DuBose T.D. Disorders of acid-base balance.in: Brenner B. Brenner & Rector's The Kidney. (ed 6). Saunders, Philadelphia, PA2008: 505-546Google Scholar:UAG=Na++K+−Cl− The normal value for UAG is −10 mEq/L (−10 mmol/L) or more negative in acidotic individuals. Urine with little or no ammonium has more combined sodium and potassium than chloride (UAG is positive), indicating a renal cause for hyperchloremic acidosis. This qualitative test is useful only in the differential diagnosis of hyperchloremic metabolic acidosis. If the patient has ketonuria or drug anions in large quantity (penicillins or aspirin) in urine, UAG is not reliable. In this situation, urinary ammonium may be estimated more reliably from the measured urine osmolality (UOSM) and other urine analytes:NH4+=0.5 (UOSM−[2 (Na++K+)+urea+glucose]) Urinary ammonium concentrations ≥75 mEq/L would be anticipated if renal tubular function is intact and the kidney is responding to the prevailing metabolic acidosis by increasing ammonium production and excretion. Conversely, values <25 mEq/L denote inappropriately low urinary ammonium concentrations. Moreover, urine pH in the patient was 6.0 (ie, >5.5) and therefore inappropriately alkaline for the prevailing systemic acidosis. Note that urine pH alone often is insufficient information or may even be misleading because patients with normal-anion-gap acidosis caused by extrarenal bicarbonate loss (diarrhea) during several days excrete large amounts of ammonium. As a buffer with a pKa of ∼9, excretion of high amounts of ammonium allows for an increase in urine pH (pH > 5.5), therefore suggesting classic distal RTA erroneously. In such cases, it often is very helpful and more accurate to estimate urinary ammonium excretion by calculation of the UAG or urine osmolar gap. The clinical response by the kidney to metabolic acidosis in this case was inappropriate, manifest by the inability to increase net acid excretion and the inability to excrete maximally acid urine (pH < 5.5). Thus, the normal-anion-gap acidosis is of renal origin (Box 2, Box 3).Box 1Diagnostic Features of Classic Distal Renal Tubular Acidosis♦Hypokalemia♦Nonanion gap metabolic acidosis♦Urine net charge (urine anion gap) is zero or positive, denoting abnormally low ammonium excretion in the face of systemic acidosis♦Urine pH > 5.5♦Modest bicarbonaturia (5% < FEHCO3 < 10%)♦Absence of Fanconi syndrome♦Abnormal calcium metabolism (hypercalciuria, nephrocalcinosis, nephrolithiasis, bone disease)♦Low urine citrate♦HyperglobulinemiaAbbreviation: FEHCO3, fractional excretion of bicarbonate.Clinical Recognition of Renal Response to Hyperchloremic AcidosisNonrenal Causes ♦Increase in net acid excretion ♦Negative urine anion gap ♦High ammonium excretion ♦Urine pH may be <5.5 or >5.5Renal Causes ♦Inability to increase net acid excretion ♦Positive urine anion gap ♦Low ammonium excretion ♦Urine pH typically >5.5 with acidosisKey Learning Points♦Serum potassium is valuable in the differential diagnosis of hyperchloremic acidosis in general and renal tubular acidosis in particular♦Calculation of urine anion and/or urine osmolal gap is useful in estimating urinary ammonium excretion to assess the contribution of the kidney to metabolic acidosis♦Urine pH can be misleading in the evaluation of renal acidification ♦Hypokalemia♦Nonanion gap metabolic acidosis♦Urine net charge (urine anion gap) is zero or positive, denoting abnormally low ammonium excretion in the face of systemic acidosis♦Urine pH > 5.5♦Modest bicarbonaturia (5% < FEHCO3 < 10%)♦Absence of Fanconi syndrome♦Abnormal calcium metabolism (hypercalciuria, nephrocalcinosis, nephrolithiasis, bone disease)♦Low urine citrate♦Hyperglobulinemia Abbreviation: FEHCO3, fractional excretion of bicarbonate. Nonrenal Causes ♦Increase in net acid excretion ♦Negative urine anion gap ♦High ammonium excretion ♦Urine pH may be <5.5 or >5.5 Renal Causes ♦Inability to increase net acid excretion ♦Positive urine anion gap ♦Low ammonium excretion ♦Urine pH typically >5.5 with acidosis ♦Serum potassium is valuable in the differential diagnosis of hyperchloremic acidosis in general and renal tubular acidosis in particular♦Calculation of urine anion and/or urine osmolal gap is useful in estimating urinary ammonium excretion to assess the contribution of the kidney to metabolic acidosis♦Urine pH can be misleading in the evaluation of renal acidification Types of renal acidosis can be characterized further by recognition of the coexistence of either hypo- or hyperkalemia (Box 4). This patient had hypokalemia and urine pH of 6 in the face of severe systemic metabolic acidosis. Thus, these features fit the classic description of classic distal RTA (type 1 RTA). This patient also did not have hyperphosphaturia or glycosuria and did not show the features of Fanconi syndrome typical of proximal renal RTA (type 2 RTA). Moreover, patients with proximal RTA often excrete relatively acid urine (pH < 5.5) in the face of metabolic acidosis (bicarbonate < 15 mEq/L [<15 mmol/L]).Types of Renal AcidosisHypokalemic forms ♦Proximal renal tubular acidosis (type 2) ♦Classic distal renal tubular acidosis (type 1)Hyperkalemic forms ♦Aldosterone deficiency or resistance (type 4) ♦Nonmineralocorticoid voltage defect ♦Uremic acidosis ♦Renal tubular acidosis of chronic kidney diseaseNormokalemic ♦Renal tubular acidosis of chronic kidney disease ♦Uremic acidosis Hypokalemic forms ♦Proximal renal tubular acidosis (type 2) ♦Classic distal renal tubular acidosis (type 1) Hyperkalemic forms ♦Aldosterone deficiency or resistance (type 4) ♦Nonmineralocorticoid voltage defect ♦Uremic acidosis ♦Renal tubular acidosis of chronic kidney disease Normokalemic ♦Renal tubular acidosis of chronic kidney disease ♦Uremic acidosis The pathogenesis of classic distal RTA in patients with Sjögren syndrome appears to be the result of an immunologic assault on the collecting tubule, causing failure to insert the adenosine triphosphate–dependent vacuolar proton pump (H+-ATPase) into the apical membrane of type A intercalated cells. Distal RTA can be the result of genetic defects in AE1 (anion exchanger 1; encoded by the SLC4A1 gene) or H+-ATPase. It has been suggested that distal RTA might be caused by a defect in H+, K+-ATPase. A single case report in a child with profound hypokalemia was assumed to have a defect in H+, K+-ATPase function.2Simpson A.G. Schwartz G.J. Distal renal tubular acidosis with severe hypokalaemia, probably caused by colonic H+-K+-ATPase deficiency.Arch Dis Child. 2001; 84: 504-507Crossref PubMed Scopus (24) Google Scholar A form of endemic distal RTA occurs with regularity in Thailand, especially in the summer. Possible causes include dietary potassium deprivation coupled with high insensible loss of potassium from sweat or as a result of environmental vanadate intoxication. Three patients have been described in whom immunocytochemical analysis of tissue obtained using kidney biopsy showed a virtual absence of H+-ATPase in the apical membrane of intercalated cells of the collecting tubule.3Poux J.M. Peyronnet P. Le Meur Y. et al.Hypokalemic quadriplegia and respiratory arrest revealing primary Sjögren syndrome.Clin Nephrol. 1992; 37: 189-191PubMed Google Scholar, 4Cohen E.P. Bastani B. Cohen M.R. et al.Absence of H+-ATPase in cortical collecting tubules of a patient with Sjögren syndrome and distal renal tubular acidosis.J Am Soc Nephrol. 1992; 3: 264-271PubMed Google Scholar, 5Bae E.H. Han C.W. Lee J.H. et al.Hypokalemia associated with nephrocalcinosis.Kidney Int. 2009; 75: 443-444Crossref PubMed Scopus (10) Google Scholar Autoantibodies against the enzyme carbonic anhydrase type II have been reported,6Takemoto F. Hoshino J. Sawa N. et al.Autoantibodies against carbonic anhydrase II are increased in renal tubular acidosis associated with Sjögren syndrome.Am J Med. 2005; 118: 181-184Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar and in a mouse model, induction of anti–carbonic anhydrase antibodies caused a urinary acidification defect.7Takemoto F. Katori H. Sawa N. et al.Induction of anti-carbonic-anhydrase-II antibody causes renal tubular acidosis in a mouse model of Sjögren syndrome.Nephron Physiol. 2007; 106: 63-68Crossref Scopus (37) Google Scholar Because this enzyme is critical for renal acid-base homeostasis,8DuBose T.D. Lucci M.S. Effect of carbonic anhydrase inhibition on superficial and deep nephron bicarbonate reabsorption in the rat.J Clin Invest. 1983; 71: 55-65Crossref PubMed Scopus (27) Google Scholar it is conceivable that a disorder affecting type II carbonic anhydrase could compromise H+-ATPase function. Alternatively, because an array of autoantibodies has been associated with systemic autoimmune diseases, such as Sjögren syndrome,9Routsias J.G. Tzioufas A.G. Sjögren syndrome—study of autoantigens and autoantibodies.Clin Rev Allergy Immunol. 2007; 32: 238-251Crossref PubMed Scopus (85) Google Scholar it is conceivable that autoantibodies against H+-ATPase could prevent trafficking and function of this transporter in the collecting tubule. Proximal RTA also has been reported in patients with Sjögren syndrome, but is much less frequent.1DuBose T.D. Disorders of acid-base balance.in: Brenner B. Brenner & Rector's The Kidney. (ed 6). Saunders, Philadelphia, PA2008: 505-546Google Scholar, 10DuBose T.D. McDonald G.A. Renal tubular acidosis.in: DuBose T.D. Hamm L. Acid-Base and Electrolyte Disorders: A Companion to Brenner and Rector's The Kidney. Saunders, Philadelphia, PA2002: 189-206Google Scholar Interestingly, in patients with primary Sjögren syndrome, the salivary gland also shows dense inflammatory infiltrate of B-type lymphocytes in areas around the ducts of the salivary gland. Although a defect in distal acidification with characteristic features of classic distal RTA is said to occur in ∼50% (18%-64%)10DuBose T.D. McDonald G.A. Renal tubular acidosis.in: DuBose T.D. Hamm L. Acid-Base and Electrolyte Disorders: A Companion to Brenner and Rector's The Kidney. Saunders, Philadelphia, PA2002: 189-206Google Scholar of patients with primary Sjögren syndrome, there is surprisingly scant information regarding the long-term follow-up of these patients to describe the expected prognosis.1DuBose T.D. Disorders of acid-base balance.in: Brenner B. Brenner & Rector's The Kidney. (ed 6). Saunders, Philadelphia, PA2008: 505-546Google Scholar, 10DuBose T.D. McDonald G.A. Renal tubular acidosis.in: DuBose T.D. Hamm L. Acid-Base and Electrolyte Disorders: A Companion to Brenner and Rector's The Kidney. Saunders, Philadelphia, PA2002: 189-206Google Scholar Nevertheless, the metabolic acidosis responds readily to alkali replacement with either sodium citrate solution or sodium bicarbonate tablets. Potassium replacement usually is not required.1DuBose T.D. Disorders of acid-base balance.in: Brenner B. Brenner & Rector's The Kidney. (ed 6). Saunders, Philadelphia, PA2008: 505-546Google Scholar One of the most frequent laboratory findings in patients with primary Sjögren syndrome with renal involvement is a high level of serum total gamma globulin.11Pertovaara M. Korpela M. Pasternack A. Factors predictive of renal involvement in patients with primary Sjögren syndrome.Clin Nephrol. 2001; 56: 10-18PubMed Google Scholar A concern in patients with Sjögren syndrome is that the tubulointerstitial infiltrate may lead to more chronic interstitial disease with tubular atrophy and fibrosis. Some patients ultimately may develop nephrocalcinosis as a result of chronic normal-anion-gap metabolic acidosis, hypercalciuria, and hypocitraturia.5Bae E.H. Han C.W. Lee J.H. et al.Hypokalemia associated with nephrocalcinosis.Kidney Int. 2009; 75: 443-444Crossref PubMed Scopus (10) Google Scholar Chronic interstitial nephritis is the major cause of progressive chronic kidney disease in patients with Sjögren syndrome. There are no studies reporting outcomes in patients receiving cytotoxic therapy and/or glucocorticoids. Recently, preliminary trials data have suggested that a decrease in B-lymphocyte infiltration in salivary gland tissue and improvement in urinary acidification occurs after treatment with rituximab.12Pijpe J. van Imhoff G.W. Spijkervet F.K. et al.Rituximab treatment in patients with primary Sjögren syndrome: an open-label phase II study.Arthritis Rheum. 2005; 52: 2740-2750Crossref PubMed Scopus (418) Google Scholar Because patients with Sjögren syndrome can develop chronic interstitial nephritis and chronic progressive kidney disease, testing the integrity of urinary acidification is recommended. If distal RTA is present, correction of metabolic acidosis with Shohl's solution or sodium bicarbonate is recommended. Support: None. Financial Disclosure: None.