Combined renal tubular acidosis and diabetes insipidus in hematological disease
Ewout J Hoorn* and Robert Zietse About the authors
Correspondence *Erasmus Medical Center, Dialysis Unit, Room Bd391, Dr Molewaterplein 40, 3015 GD Rotterdam, The Netherlands
Email ejhoorn@gmail.com
Summary
Background A 39-year-old male with multiple myeloma was admitted for treatment with melphalan and autologous stem cell reinfusion. He presented with hypokalemia and hyperchloremic non-anion-gap metabolic acidosis with a high urinary pH. He also had hypomagnesemia, hypophosphatemia, hypouricemia, proteinuria and glucosuria. The patient subsequently developed polyuria with a low urine osmolality, hypernatremia and, finally, acute renal failure.
Investigations Physical examination, blood and urine analyses, kidney biopsy and tonicity balance.
Diagnosis Fanconi syndrome with proximal (type II) renal tubular acidosis caused by myeloma kidney. Renal tubular acidosis was complicated by probable nephrogenic diabetes insipidus and acute renal failure.
Management Potassium supplementation, sodium bicarbonate therapy, intravenous fluid therapy and dialysis.
Keywords:
amphotericin B, Fanconi syndrome, hypernatremia, hypokalemia, multiple myeloma
The case
A 39-year-old male with stage II IgG-
multiple myeloma was admitted to hospital for melphalan chemotherapy and autologous stem cell reinfusion. Multiple myeloma had been diagnosed 6 months previously when the patient had presented with tetraparesis resulting from a metastasis in the second cervical vertebra. Apart from the tetraparesis, the patient's history and physical examination were unremarkable. Blood and urine tests at admission revealed the following immunological parameters: serum 
2 microglobulin level 3.93 mg/l, serum IgG- M-component 15 g/l, serum light chain level 2,650 mg/l and urinary light chain level 2.97 g/l.
The patient presented with metabolic acidosis (serum pH 7.25, urine pH 7) and a normal anion gap (serum sodium 151 mmol/l, chloride 125 mmol/l, bicarbonate 15.9 mmol/l; anion gap 10 mmol/l). He also had the following biochemical abnormalities: hypokalemia (3.0 mmol/l), hypomagnesemia (0.63 mmol/l), hypouricemia (0.08 mmol/l), hypophosphatemia (0.77 mmol/l), glucosuria (with a normal serum glucose level) and proteinuria (0.28 g/day). The patient was not taking diuretics and he did not have gastrointestinal losses or hypertension. Sodium bicarbonate treatment corrected his acidosis, but resulted in a further fall in serum potassium level (to 2.6 mmol/l), which was treated with potassium supplementation.
Over the next 2 days, the patient developed polyuria (4–7 l urine per day) with a urine osmolality of 99 mmol/kg and a urinary sodium concentration of 7 mmol/l. Although the presence of diabetes insipidus was not confirmed by vasopressin analog administration, there were no other obvious reasons for polyuria as glucose, urea and mannitol levels were normal. Polyuria resulted in a negative fluid balance and a further increase in serum sodium concentration to 158 mmol/l, as shown in a so-called 'tonicity balance' (Figure 1A).
Figure 1 Tonicity balances showing two different mechanisms of hypernatremia
The final 24-hour rise in serum sodium concentration is shown for (A) the main case and (B) for the case described in Box 1. The large central rectangle in each diagram represents total body water with the serum sodium concentration measured at the start and end of the observation period shown above and below this rectangle, respectively. The quantities of Na+ plus K+ infused and excreted are shown in the two flanking rectangles, and the volumes of water infused and excreted are depicted below the dashed line. Abbreviation: SNa, serum sodium concentration.
Full figure and legend (22K)Figures & Tables indexDownload Power Point slide (132K)On the third day of hospitalization the patient also developed acute renal failure, initially with a serum creatinine level of 150 
mol/l (1.7 mg/dl) and fractional excretions of 0.9% for sodium and 81.0% for potassium. Over the next 5 days, the patient's serum creatinine level increased to a maximum of 738 mol/l (8.3 mg/dl). A kidney biopsy was performed at day 8 (Figure 2). The patient was diagnosed with Fanconi syndrome with proximal (type II) renal tubular acidosis (RTA) caused by myeloma kidney. He was started on hemodialysis, which partly restored renal function, but 2 months later the patient is still being seen as an outpatient (his current serum creatinine level is approximately 150 mol/l [1.7 mg/dl] and he does not require dialysis).
Figure 2 Kidney biopsy of the main case
(A) Kidney biopsy taken during renal failure and showing tubular casts (arrow), interstitial fibrosis, infiltration with lymphocytes and intact glomeruli. (B) A detailed image showing a tubular cast (solid arrow) with a ring of macrophages (dashed arrows).
Full figure and legend (63K)Figures & Tables indexDownload Power Point slide (172K)We also report a case of distal (type I) RTA and nephrogenic diabetes insipidus (NDI) caused by amphotericin B treatment (Box 1).
Box 1 A case of distal (type I) renal tubular acidosis and nephrogenic diabetes insipidus caused by amphotericin B treatment.
A 34-year-old male with a congenital syndrome characterized by partial bone marrow insufficiency developed myelodysplastic syndrome (type: refractory anemia with excess blasts). He was admitted to hospital for chemotherapy with cytarabine.
Treatment of the patient's myelodysplastic syndrome was complicated by a fungal infection (Aspergillus fumigatus) of the jaw for which he received amphotericin B in a lipid-based formulation. On the third day of hospitalization, the patient developed hypokalemia (serum potassium 2.1 mmol/l) and metabolic acidosis (serum pH 7.05, urine pH 7) with a normal anion gap (serum sodium 147 mmol/l, chloride 128 mmol/l, bicarbonate 6.9 mmol/l; anion gap 12 mmol/l). Calcium, urate and phosphate levels were within normal ranges. The patient was not using diuretics, had no gastrointestinal loss, and no hypertension. Treatment consisted only of potassium supplementation, which slowly corrected the hypokalemia over 8 days.
Five days after starting amphotericin B, the patient developed polyuria (maximum 7.7 l urine per day) with a low urine osmolality (267 mmol/kg). Intravenous administration of desmopressin acetate (3 g), a vasopressin analog, resulted in a minimal increase in urine osmolality to 300 mmol/kg. Treatment with large volumes of isotonic intravenous fluids resulted in severe hypernatremia (peak serum sodium 171 mmol/l), as depicted in a tonicity balance (Figure 1B). Finally, during the ninth and final day of amphotericin B treatment, acute renal failure (peak serum creatinine 188 mol/l [2.1 mg/dl]) developed with concomitant hyperkalemia (serum potassium 7.4 mmol/l) during continued potassium supplementation. Given the poor prognosis, abstinence from therapy was agreed and the patient died shortly after.
Discussion of diagnosis
Both cases presented with hypokalemia and a normal anion gap metabolic acidosis without evidence for gastrointestinal bicarbonate loss. As both patients also had a relatively high urine pH, RTA seemed likely in both cases. The main case described showed additional signs of proximal tubular dysfunction (hypouricemia, hypophosphatemia, glucosuria and proteinuria), indicating a diagnosis of Fanconi syndrome. The patient described in Box 1 probably had distal RTA, not only because proximal tubular dysfunction was absent (no hypouricemia, hypophosphatemia, glucosuria, or proteinuria), but also because his serum bicarbonate level was less than 10 mmol/l.1
Shortly after development of RTA, both patients developed polyuria with a low urine osmolality and hypernatremia. As there was no evidence for osmotic diuresis, polydipsia or central diabetes insipidus (CDI) in either patient, NDI seemed likely in both patients; however, it was confirmed only in the patient described in Box 1. As confirmation of NDI was not pursued in the main patient described, he could theoretically have had CDI, but there are no examples in the literature to support an association between multiple myeloma and CDI. The high osmole excretion rate in the patient described in Box 1 (Figure 1B) indicates that saline diuresis might also have contributed to polyuria.
Although the available parameters support our diagnoses, we acknowledge that more-conclusive diagnostic tests might have been useful; for example, the measurement of fractional bicarbonate excretion during bicarbonate infusion, and the measurement of other urinary parameters such as calcium, magnesium, phosphate, ammonium, partial carbon dioxide pressure and citrate levels.1, 2 The majority of these tests are, however, not available at our institution.
Discussion of possible mechanisms of tubular dysfunction
The fact that both patients probably had a combination of RTA and NDI led us to consider the mechanisms that might be involved in compromising renal tubular transport, and how the coexistence of the two conditions might be explained.
Relevant to the main case presented, Messiaen et al. showed that Fanconi syndrome secondary to multiple myeloma might be related to the presence of proteolysis-resistant -type light chains, which can accumulate in proximal tubule cells and impair their function.3 By contrast, the mechanism of NDI in multiple myeloma is less clear.4 We do know, however, that once proximal tubular dysfunction develops, delivery of light chains to the distal nephron increases, leading to dysfunction, particularly in the loop of Henle. One possible mechanism that might link Fanconi syndrome with NDI, therefore, is that loop of Henle dysfunction interferes with the kidney's concentrating mechanism. Although inconclusive, the presence of hypomagnesemia and an extremely high fractional excretion of potassium (as occurred in the main case) might indicate loop of Henle dysfunction.5 This dysfunction might result from tubular cast formation (Figure 2), which occurs in this nephron segment if light chains react with Tamm–Horsfall protein. In 1990, Sanders et al. showed that cast-forming proteins aggregated with Tamm–Horsfall protein in vitro and caused loop of Henle dysfunction in rats in vivo that was characterized by decreased chloride reabsorption.6 In 1999, Matsumura et al. described another link between loop of Henle dysfunction and a reduction in chloride reabsorption by showing that mice lacking the CLC-K1 chloride channel in the loop of Henle developed NDI.7 Finally, inactivation of Tamm–Horsfall protein by cast formation might also contribute to loop of Henle dysfunction, as was recently demonstrated in Tamm–Horsfall knockout mice.8
The patient presented in Box 1 developed distal RTA and NDI during treatment with amphotericin B. Amphotericin B can cause NDI by impairing the expression of aquaporin-2 water channels through an effect on adenylyl cyclase.9 We speculate that any relationship between distal RTA and NDI in the patient described in Box 1 might also be explained by an effect of amphotericin B on adenylyl cyclase. Namely, adenylyl cyclase V regulates vacuolar H+ ATPase, which is the main transporter responsible for H+ ion secretion in the renal collecting duct.10 It is believed, however, that amphotericin B causes RTA by increasing membrane permeability in the collecting duct, resulting in a back-flux of H+ ions.1 We might have been able to determine which particular mechanism of distal RTA was present in the case described in Box 1 by assessing the response of the urine-to-blood carbon dioxide pressure index to the administration of bicarbonate.2
Of final consideration in both cases presented is the role of hypokalemia. In animals, hypokalemia can induce NDI by a downregulation of aquaporin-2 water channel expression secondary to decreased cyclic AMP production,11 although reduced expression of additional renal transporters further upstream might also contribute.12 Chronic hypokalemia is also known to induce morphological kidney changes such as interstitial fibrosis, which was seen in the kidney biopsy of the main patient described (Figure 2).1 Using a large number of human kidney biopsies, Bedford et al. showed that aquaporin-2 expression was lower in diseased kidneys than in healthy kidneys, particularly in lesions with interstitial fibrosis.13 Interestingly, amphotericin B can also induce interstitial fibrosis.14 It is not known, however, whether the presence of these morphological changes directly impairs tubular transport, or whether impaired transport is merely a secondary phenomenon.
Discussion of management
The two cases described illustrate the clinical challenges that can arise from tubular dysfunction, such as disorders of potassium and sodium balance, and acute renal failure.
Alkali therapy can worsen hypokalemia in a patient with proximal (type II) RTA as it can cause a bicarbonate diuresis with subsequent potassium wasting. This outcome might have been prevented in the main case described because alkali therapy was given in the form of a potassium salt.1 Conversely, alkali therapy can effectively correct hypokalemia in distal RTA,1 as shown in the case described in Box 1. Given that this patient went on to develop hyperkalemia, however, potassium supplementation should be titrated according to renal function.
Hypernatremia is a potentially dangerous but avoidable complication of inadequate intravenous fluid therapy. In the main case described, a negative water balance contributed to hypernatremia, probably because the ongoing water diuresis was compensated for with insufficient intravenous fluids. Conversely, in the patient described in Box 1, sodium balance was more positive than fluid balance because excretion of large volumes of hypotonic urine was compensated for by even larger volumes of predominantly isotonic intravenous fluids. To analyze the cause of hypernatremia, we used a 'tonicity balance', in which separate mass balances are calculated for sodium plus potassium and for water.15 In order to prevent hypernatremia during NDI, the concept of infusing 'isotonic to patient', meaning that the tonicity of the infusate should equal that of the produced urine, is important.
The negative water balance caused by NDI in the main case described might also have contributed to acute renal failure, as the low fractional sodium excretion suggested a prerenal origin. Cast formation can also be stimulated by volume depletion and might, therefore, have been an additional trigger for the deterioration in renal function.6 In the case described in Box 1, amphotericin B was the most likely culprit for acute renal failure even though it was given as a lipid complex, which is believed to be less nephrotoxic than other formulations.1
Conclusions
Several diseases and drugs can perturb renal tubular function in different nephron segments, potentially resulting in acid–base or electrolyte disorders. We have described two patients with hematological disease who developed combined RTA and NDI. RTA should be suspected in all patients who present with a non-anion-gap metabolic acidosis, hypokalemia, and a high urinary pH, in the absence of gastrointestinal bicarbonate loss. Diagnosing RTA and its type is important because sodium bicarbonate therapy might either be indicated (distal RTA) or potentially harmful (proximal RTA) in the management of hypokalemia. NDI should be suspected when vasopressin-resistant polyuria with a low urinary osmolality develops. Treatment of NDI requires careful and tailored intravenous fluid therapy to prevent hypernatremia caused by a negative water balance or a positive sodium balance. A tonicity balance might prove to be a useful bedside tool to differentiate hypernatremic disorders and to help organize their treatment.15 Finally, Table 1 summarizes diseases and drugs that can cause combined RTA and NDI. It is important for clinicians to be aware of these possible concurrences and to monitor patients accordingly.
Table 1 Diseases and drugs than can induce combined tubular dysfunctiona
Full tableFigures & Tables indexDownload Power Point slide (164K)
Acknowledgments
We thank Dr IM Bajema, who analyzed the kidney biopsy, Dr B van den Berg and Dr MR Korte, who were also involved in the treatment of these patients, and Dr ML Halperin for critical reading of this manuscript.
References
- Rose BD and Post TW (2001) Clinical Physiology of Acid–Base and Electrolyte Disorders, edn 5. New York: McGraw-Hill
- Stinebaugh BJ et al. (1981) Pathogenesis of distal renal tubular acidosis. Kidney Int 19: 1–7 | Article | PubMed | ChemPort |
- Messiaen T et al. (2000) Adult Fanconi syndrome secondary to light chain gammopathy: clinicopathologic heterogeneity and unusual features in 11 patients. Medicine (Baltimore) 79: 135–154 | Article | PubMed | ChemPort |
- Smithline N et al. (1976) Light-chain nephropathy: renal tubular dysfunction associated with light-chain proteinuria. N Engl J Med 294: 71–74 | PubMed | ChemPort |
- Elisaf M and Siamopoulos KC (1995) Fractional excretion of potassium in normal subjects and in patients with hypokalaemia. Postgrad Med J 71: 211–212 | PubMed | ChemPort |
- Sanders PW et al. (1990) Mechanisms of intranephronal proteinaceous cast formation by low molecular weight proteins. J Clin Invest 85: 570–576 | Article | PubMed | ChemPort |
- Matsumura Y et al. (1999) Overt nephrogenic diabetes insipidus in mice lacking the CLC-K1 chloride channel. Nat Genet 21: 95–98 | Article | PubMed | ISI | ChemPort |
- Bachmann S et al. (2005) Renal effects of Tamm–Horsfall protein (uromodulin) deficiency in mice. Am J Physiol Renal Physiol 288: F559–F567 | Article | PubMed | ISI | ChemPort |
- Kim SW et al. (2001) Amphotericin B decreases adenylyl cyclase activity and aquaporin-2 expression in rat kidney. J Lab Clin Med 138: 243–249 | Article | PubMed | ChemPort |
- Héliès-Toussaint C et al. (2000) Cellular localization of type 5 and type 6 ACs in collecting duct and regulation of cAMP synthesis. Am J Physiol Renal Physiol 279: F185–F194 | PubMed |
- Marples D et al. (1996) Hypokalemia-induced downregulation of aquaporin-2 water channel expression in rat kidney medulla and cortex. J Clin Invest 97: 1960–1968 | PubMed | ISI | ChemPort |
- Elkjær M-L et al. (2002) Altered expression of renal NHE3, TSC, BSC-1, and ENaC subunits in potassium-depleted rats. Am J Physiol Renal Physiol 283: F1376–F1388 | PubMed | ISI | ChemPort |
- Bedford JJ et al. (2003) Aquaporin expression in normal human kidney and in renal disease. J Am Soc Nephrol 14: 2581–2587 | Article | PubMed | ChemPort |
- Reynolds ES et al. (1963) The renal lesion related to amphotericin B treatment for coccidioidomycosis. Med Clin North Am 47: 1149–1154 | PubMed | ChemPort |
- Carlotti AP et al. (2001) Tonicity balance, and not electrolyte-free water calculations, more accurately guides therapy for acute changes in natremia. Intensive Care Med 27: 921–924 | PubMed | ChemPort |
- Minemura K et al. (2001) IgA-kappa type multiple myeloma affecting proximal and distal renal tubules. Intern Med 40: 931–935 | PubMed | ChemPort |
- Spruce BA et al. (1984) Idiopathic hypergammaglobulinaemia associated with nephrogenic diabetes insipidus and distal renal tubular acidosis. Postgrad Med J 60: 493–494 | PubMed | ChemPort |
- Folami AO et al. (1978) Coexistence of distal renal tubular acidosis and nephrogenic diabetes insipidus in two patients: implications for the pathogenesis of distal renal tubular acidosis. Clin Invest Med 1: 105–109 | PubMed | ChemPort |
- Nagayama Y et al. (1994) Acquired nephrogenic diabetes insipidus secondary to distal renal tubular acidosis and nephrocalcinosis associated with Sjögren's syndrome. J Endocrinol Invest 17: 659–663 | PubMed | ChemPort |
- Vigeral P et al. (1987) Nephrogenic diabetes insipidus and distal tubular acidosis in methicillin-induced interstitial nephritis. Adv Exp Med Biol 212: 129–134 | PubMed | ChemPort |
- Navarro JF et al. (1996) Nephrogenic diabetes insipidus and renal tubular acidosis secondary to foscarnet therapy. Am J Kidney Dis 27: 431–434 | PubMed | ChemPort |
- Negro A et al. (1998) Ifosfamide-induced renal Fanconi syndrome with associated nephrogenic diabetes insipidus in an adult patient. Nephrol Dial Transplant 13: 1547–1549 | PubMed | ISI | ChemPort |
- Karras A et al. (2003) Tenofovir-related nephrotoxicity in human immunodeficiency virus-infected patients: three cases of renal failure, Fanconi syndrome, and nephrogenic diabetes insipidus. Clin Infect Dis 36: 1070–1073 | Article | PubMed | ISI |
- Rollot F et al. (2003) Tenofovir-related Fanconi syndrome with nephrogenic diabetes insipidus in a patient with acquired immunodeficiency syndrome: the role of lopinavir-ritonavir-didanosine. Clin Infect Dis 37: e174–e176 | Article | PubMed | ISI |
Competing interests
The authors declared no competing interests.
Contact the journal about this article
Subject areas under which this article appears: Acid-base, fluid and electrolyte disorders | Acute renal failure
