Article Text

Neurogenic diabetes insipidus in patients with systemic lupus erythematosus
  1. JULIO SÁNCHEZ-ROMÁN,
  2. MARÍA JESÚS CASTILLO-PALMA,
  3. CELIA OCAÑA MEDINA
  1. Unidad de Colagenosis
  2. Department of Endocrinology
  3. Department of Immunology, Hospital Universitario Virgen del Rocío, Seville, Spain
  1. Dr J Sánchez-Román, Plaza de San Martín, 3 F, 2ª 41003, Seville, Spain.
  1. FERNANDO VILLAMIL-FERNÁNDEZ
  1. Unidad de Colagenosis
  2. Department of Endocrinology
  3. Department of Immunology, Hospital Universitario Virgen del Rocío, Seville, Spain
  1. Dr J Sánchez-Román, Plaza de San Martín, 3 F, 2ª 41003, Seville, Spain.
  1. INGEBORG WICHMANN
  1. Unidad de Colagenosis
  2. Department of Endocrinology
  3. Department of Immunology, Hospital Universitario Virgen del Rocío, Seville, Spain
  1. Dr J Sánchez-Román, Plaza de San Martín, 3 F, 2ª 41003, Seville, Spain.

Statistics from Altmetric.com

Neurogenic diabetes insipidus (NDI) approximately affects one of 25 000 people. The idiopathic type represents about 30% of all NDI cases (one of 75 000).1 The coexistence of primary NDI and systemic lupus erythematosus (SLE) is an exceptional fact that has not been previously described. We report two cases belonging to a series of 350 SLE patients (ARA criteria,1988).2

CASE 1

A 48 year old woman who had suffered from pulmonary tuberculosis in 1976. Ten years later she was examined because of a suspected myopericarditis. She also had a photosensitive malar erythema, oral ulcers, Raynaud’s phenomenon, vasculitic lesions on fingertips, Sjögren’s syndrome, and polyarthritis during one year. Blood count, complement, protein electrophoresis, rheumatoid factor, syphilis serological test, and renal function were normal. Antinuclear antibodies (ANA) were positive (1/160). Anticardiolipin antibodies (ACA) were negative. Lupus anticoagulant test (LA) only showed once a prolonged Russell test at the limit of significance (34.8 s). The patient had polyuria-polydipsia for an undetermined time. NDI was detected by using the thirst test (fig 1). Remaining pituitary functions integrity was demonstrated by basal and dynamic hormonal quantifications (table 1). Antihypophysis antibodies (monkey hypophysis substrate, Sanofi Pasteur, Marnes-La-Coquette) were negative. Sellar and parasellar areas (computed tomography and magnetic imaging) were normal. The polyuria-polydipsia syndrome was successfully controlled with substitutive hormonal treatment (ADH).

Figure 1

Thirst test. Osmolality (urine). *= administration of DDAVP.  

Table 1

Pituitary evaluation

CASE 2

A 69 year old man who had suffered deep venous thrombosis (DVT) on his left leg in 1983, and another episode a year later. Four years later, he was admitted to hospital because of a longlasting fever. He presented with polyarthritis, leucolymphopenia, thrombopenia, and ANA positive test (1/1280). Glucocorticoids were administered and fever disappeared. Three years later he suffered a new DVT on his right leg and, one year later, a pulmonary thromboembolism. Raised ACA (45 GPL, 20 MPL) were detected and coagulation and LA confirming tests were normal. He also suffered a psychotic episode requiring hospitalisation in 1991. Four years later, endocrinological tests were performed because of recent development of polyuria and impotency. The thirst test showed the diabetes insipidus typical pattern (fig 1). Testosterone, LH, FSH, and GH basal concentrations were below normal values. The GH response to insulin induced hypoglycaemia and the gonadotrophine response to TRH and LHRH were low. The hormonal study excluded hypothalamus-hypophysis-adrenal axis and hypothalamus-hypophysis-thyroid axis alterations (table 1). Antihypophysis antibodies were negative. Computed tomography and magnetic resonance imaging of sellar and parasellar areas were normal. The patient did well with antiplatelet and substitutive hormonal treatment (ADH and testosterone).

The frequency of NDI in our patients with SLE (0.57%), though relatively low, cannot be explained only at random as the stimated NDI frequency in the general population is 0.0013%. The theoretic probability to observe two cases in a random population of 350 cases would be 0.00108% (Poisson distribution), which is much less than the observed frequency in the patient series (p<0.0000001). No mention was found that related NDI with SLE patients in the data base MEDLINE (1992–1997) and in specialised publications related to SLE.3 4 Some isolated NDI cases associated with vasculitis have been reported (Wegener’s granulomatosis5or temporal arteritis6), so it is possible that the patients underlying lesion described in this paper could be because of a vascular impairment of the hypothalamus or hypophysis stalk. Some authors have suggested a relation between adrenal insufficiency resulting from glandular infarctions and ACA in patients with or without SLE.7-9 It could be that the patients presented here had vascular impairment, not because of inflammatory causes, but because of thrombotic events. Finally we cannot rule out a direct autoimmune lesion. Schgeerbaum and Botazzo10published the existence of autoantibodies against vasopressin producing neurons in some patients suffering from NDI. However, the antibodies against hypophysis structures were negative by IIF in our two cases.

References

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.