Aquaporins and the surgeon

Nephrogenic diabetes insipidus (NDI) presents an uncommon but formidable clinical challenge in the surgical patient. Two recent cases of NDI with differing aetiology are presented. These cases and a review of the literature illustrate well the diagnosis, fluid and electrolyte imbalances seen and the strategy of treatment required in the post-operative setting. The central role of the recently discovered aquaporin channels in this condition is briefly outlined. Nephrogenic diabetes insipidus has a diverse aetiology and many of the hazards of the condition are peculiar to the surgical setting. The importance of management in a high dependency environment is highlighted.

Keywords: Aquaporin, high dependency unit, hypercalcaemia, lithium, nephrogenic diabetes insipidus, surgery

A 47-year-old man was admitted to the high dependency unit (HDU) following emergency laparotomy. He had a life-long history of severe hidradenitis suppuritiva complicated by chronic sepsis and anaemia but otherwise no significant medical problems. Whilst being investigated for worsening health and enlargement of an area of hydradenitis in the posterior neck, he developed severe abdominal pain, leucocytosis, pyrexia and hypotension. At laparotomy, a splenic abscess was found and splenectomy performed. Biopsy of the neck lesion was carried out concurrently.

In the immediate post-operative period it was noted that the plasma sodium concentration was 156.9 mmol/l and a serum osmolarity of 330 mOsmol/l. However, he passed 8 litres of dilute urine in 24 hours (<200 mOsmol/l) despite a central venous pressure (CVP) of -1 cm H2O. A diagnosis of diabetes insipidus was thought likely, fluids were actively replaced, as guided by the CVP and serum sodium, and an aetiology sought. Suspicion of malignant change in the neck lesion was high, and serum calcium was found to be elevated at 4.15 mmol/l (adjusted).A diagnosis of nephrogenic diabetes insipidus (NDI) secondary to hypercalcaemia was made. Parathyroid hormone levels were found to be normal and biopsies of the neck lesion proved to be an anaplastic carcinoma.

A 66-year-old woman was admitted to the HDU following anterior resection of a locally advanced (T4NO) rectal adeno-carcinoma. She had presented with weight loss and diarrhoea and, at operation, a pelvic abscess was demonstrated with tumour involving the bladder, uterus and small bowel. Her past medical history was notable only for schizophrenia with depression.

In the first post-operative day urine output was 5.4 litres, and a serum sodium of 157.4 mmol/l and a chloride of 124.0 mmol/l was recorded. Over the next few days, urine volumes peaked at 11 litres and the serum sodium at 161.0 mmol/l, whilst serum and urine osmolarity were 338 and 202 mOsmol/l, respectively. Nephrogenic diabetes insipidus secondary to the effects of lithium was thought likely and whilst pre-operative lithium levels were within the therapeutic range, no other possible aetiology was identified.

Hydradenitis suppuritiva (HS) is a recurrent, suppurative disease manifested by abscesses, fistulas and scarring. Malignant change within HS is a rare and serious consequence, usually occurring in the groin and anogenital region in long standing cases.1 Hypercalcaemia, as a complication of such malignant change in the groin, has been reported on nine occasions in the literature and has been shown to be associated with elevated parathyroid hormone-related protein, as seen in squamous cell carcinoma of the lung. 2,3 Previous cases in the neck region have not been described.

Hypercalcaemia is a well described cause of NDI although its mechanism has only recently been elucidated.4 Aquaporins were first described by Agre et al (1993) and are a family of water channel proteins which provide a pathway for water transport across cell membranes.5 At least six aquaporins are now known to be expressed in the kidney.6 The predominant vasopressin regulated water channel, known as aquaporin-2 (AQP2) is known to be involved in the pathophysiology of NDI. The mechanism of signalling cascades between hypercalcaemia and decreased AQP2 expression are only partly understood but may be mediated via lower cAMP production and increased prostaglandin E2 synthesis. Clearly hypercalcaemia is a common disorder associated with malignancy, hyperparathyroidism and certain drugs. These causes should be considered when NDI is thought possible.

Lithium-induced NDI is relatively common, affecting renal concentrating ability in up to 30%, and causing frank NDI in up to10% of patients.7 Its mechanism, obscure until recently, is now known to involve reduced expression of AQP-2 at the transcriptional level.8 Significantly, the renal effects of lithium may continue for many weeks after the drug is stopped and, in one case report, 8 years.9 The hazards of lithium-related NDI, consequent dehydration and hypernatraemia in the post-operative period are relatively poorly documented in the literature, only one case report being revealed in a Med Line search. 10

The various causes of NDI are presented in Table 1 (cranial diabetes insipidus is a distinct entity and may be caused by cranial trauma, neoplasia or infection). Nephrogenic diabetes insipidus is characterised by resistance of the kidney to vasopressin, and vasopressin analogues have no role in the management of NDI. Congenital NDI is thought to be caused by mutations in the genes coding for AQP2 and all known acquired mechanisms involve decreased expression of AQP2 in the renal tubule.8

Congenital	Aquired

X-linked recessive NDI	Metabolic - hypercalcaemia, hypokalaemia
	Autosomal recessive NDI
	Renal - chronic renal failure, amyloid, pyelonephritis, sarcoid
	Drugs - lithium, amphotericin, glibenclamide, demeclocycline
	Osmotic - glucose, post-obstructive uropathy
	Vascular - sickle cell disease
	Idiopathic

Nephrogenic diabetes insipidus can present a formidable clinical problem in the peri-operative period, if not recognised and treated actively. The inappropriate excretion of large volumes of dilute urine will leave the patient dehydrated and hypernatraemic, with increasing coma, organ hypo-perfusion and eventual death. The risks are compounded at times of restricted oral fluid intake, intra-venous fluids, acute illness and exaggerated fluid losses - a not uncommon situation following major abdominal surgery. Use of urine volumes as a guide for fluid resuscitation in the post-operative period is hazardous in these patients.

Diagnosis of NDI is reliant on careful peri-operative monitoring of fluid balance and electrolyte measurements as well as a high index of clinical suspicion.Typically, the diagnosis cannot be made without a formal water deprivation test. However, when possible, this is likely to result in unacceptable deterioration in the post-operative patient. The measurement of urine volumes, CVP, serum sodium/ osmolality and urine osmolality in the presence of a known aetiological factor are sufficient to commence treatment. Interestingly, hypernatraemia has been previously identified as a marker of both inadequate treatment and poor outcome in the critical care environment. 11

Treatment of NDI is directed towards two goals. Firstly, the underlying cause should be identified and eliminated. In Case 1, described above, hypercalcaemia was corrected by giving intravenous pamidronate, which although having a delayed effect, was eventually effective. In Case 2, the NDI persisted despite stopping lithium, as is well described in the literature. Pharmacological measures have been described such as giving thiazide and/or amiloride diuretics or more recently high dose prostaglandin sythetase inhibitors (e.g. indomethacin). 12 These methods have some evidence of efficacy but consequent risks of inducing renal failure make involvement of a renal physician desirable.

Secondly, the fluid and electrolyte imbalances must be corrected. The principle of replacement of a deficit, maintenance and any on-going losses of water and electrolytes is often a useful guide, but caution must be exercised to avoid cerebral oedema by over-rapid correction of hypernatraemia. In severe cases (Na >170mmol/l), initial use of 0.9% saline should allow sodium to fall over 48 hours. In less severe cases (Na 150-170 mmol/l) it is safe to use hypotonic solutions such as 0.45% saline or 5% glucose.4 In the cases above, hourly urine volumes were added to drain, stoma and insensible losses and replaced with 5% dextrose until serum sodium levels returned to the reference range. Volume replacement should also be guided by CVP measurements, daily weightings and frequent clinical examination. With resolution of ileus, oral intake is allowed and the patient’s own homeostatic mechanisms aid fluid management. Thus, intravenous fluids can gradually be weaned provided the patient has free access to drinking water and thirst is not obtunded by a clouded conscious state. Clearly, the successful diagnosis and treatment of NDI in a peri-operative setting necessitates an intensive care approach, such as can be offered in an HDU.

The author would like to thank Mr Andrew McEwen and Mr Colin Porteous (Consultant Surgeons, RAH Paisley) for allowing their patient to be the subject of this case report. Thanks is also expressed to the nursing staff in the HDU for their central role in the management of these patients.

1. Brown TJ, Rosen T, Orengo IF. Hidradenitis suppurativa. South Med J 1998; 91:1107-14
2. Sparks MK, Kuhlman DS, Prieto A, Callen JP. Hypercalcaemia in association with cutaneous squamous cell carcinoma. Arch Dermatol 1985;121: 243-6
3. Welsh DA, Powers JS. Elevated parathyroid hormone-related protein and hypercalcaemia in a patient with cutaneous sqaumous cell carcinoma complicating hidradenitis suppuritiva. South Med J 1993; 86:1403-4
4. Weatherall DJ, Ledingham JGG, Warrell DA. Oxford Textbook of Medicine. Oxford: Oxford University Press, 1996: 3120
5. Agre P, Sasaki S, Chrispeels MJ. Aquaporins: a family of water channel proteins. Am J Physiol 1993; 265: F461
6. Nielsen S, Kwon TH, Christensen BM, Promeneur D, Frokiaer J, Marples D. Physiology and pathophysiology of renal aquaporins. J Am Soc Nephrol 1999; 10: 647-63
7. Earm JH, Christensen BM, Frokler J, Marples D, Han J, Knepper MA, Nielsen S. Decreased aquaporin-2 expression and apical plasma membrane delivery in kidney collecting ducts of polyuric hypercalcaemic rats. J Am Soc Nephrol 1998; 9: 2181-93
8. Hyperosmolar coma due to lithium-induced diabetes insipidus. Lancet 1995; 346: 413-7
9. Oksche A, Rosenthal W. The molecular basis of nephrogenic diabetes insipidus. J Mol Med 1998; 76: 326-37
10. Stone KA. Lithium-induced nephrogenic diabetes insipidus. J Am Board Fam Pract 1999;12: 43-7
11. Johnson MA, Ogorman J, Golembiewski GH, Paluzzi MW. Nephrogenic diabetes insipidus secondary to lithium therapy in the post-operative patient: a case report. Am Surg 1994; 60: 836-9
12. Polderman KH, Schreuder WO, Strack RJ, Thijs LG. Hypernatraemia in the intensive care unit: an indicator of quality of care? Crit Care Med 1999; 27: 1105-8
13. Lam SS, Kjellstrand C. Emergency treatment of lithium induced diabetes insipidus with NSIADS. Ren Fail 1997; 19:183-8

Copyright date: 27th March 2001
Correspondence: Mr RJ Shaw, Maxillofacial Unit, University Hospital Aintree, Longmoor Lane, Liverpool L9 7AL, UK
E-mail: richardjohnshaw@hotmail.com

Aquaporins and the surgeon: cautionary tales