Nephrogenic Diabetes Insipidus Secondary to Lithium Therapy in the Postoperative Patient: A Case Report
| Title: | Nephrogenic Diabetes Insipidus Secondary to Lithium Therapy in the Postoperative Patient: A Case Report |
|---|---|
| Authors: | Johnson, M.D., Michael A.; Ogorman, M.D., Joseph; Golembiewski, M.D., Geoffrey H.; Paluzzi, MD, Michael W. |
| Publisher: | American Surgeon |
| Date Published: | November 01, 1994 |
| Reference Number: | 87 |
Nephrogenic diabetes insipidus (NDI) presents a rarely encountered but challenging fluid management problem in the perioperative period. This case is that of a patient with a perforated duodenal ulcer and previously undiagnosed NDI who received standard preoperative and postoperative hydration with normal saline, causing hypernatremia and an inappropriate diuresis. The resulting hypernatremia and hyperosmolality required aggressive hypotonic fluid replacement to return to preoperative values. Though refractory to 1-desamino-8-D-argenine-vasopressin (dDAVP), thiazide diuretics and nonsteroidal anti-inflammatory agents have a role in managing selected patients. Early diagnosis with careful fluid and electrolyte management are critical in successful management of these patients in the perioperative period.
Nephrogenic diabetes insipidus is a well known complication of long term lithium therapy1 occurring in 12 per cent to 33 per cent of the patients treated with lithium.2 The syndrome is characterized by a water diuresis of variable severity in the face of high AVP levels, and normal or low intravascular filling pressures. The disorder is well tolerated because of the extreme sensitivity of the thirst mechanism in maintaining plasma osmolality, but can cause major morbidity in conditions where oral fluid intake is curtailed. This disorder is often irreversible and without a completely effective practical treatment. The following is a report of a patient with previously undiagnosed nephrogenic diabetes insipidus (NDI) secondary to lithium who required anesthesia, surgery, and oral fluid deprivation because of a perforated duodenal ulcer.
The patient is a 42-year-old male with a history of schizophrenia and mild hypertension who presented to the emergency room with a 24-hour history of abdominal pain, nausea, vomiting, and anorexia. Physical examination revealed diffuse abdominal tenderness, with guarding and rebound tenderness. Radiographic evaluation revealed a moderate amount of subdiaphragmatic free air on the PA chest and upright films. Laboratory studies of note are as follows: serum sodium 130 meq/L (137-147), chloride 89 meq/L (94-106), creatinine 2.0 mg/dL (0.7-1.3), BUN 27 mg/dL (7-21).
Fluid resuscitation was initiated with normal saline, and the patient was prepared for surgery. Findings during exploration included a perforated duodenal ulcer with significant inflammation of the antro-duodenal region and moderate contamination of the peritoneal cavity. A Grahm patch closure was performed and a feeding jejunostomy was placed. The patient's early postoperative course was remarkable for a gradual increase in urine output that was initially thought to be related to overly aggressive resuscitation. By approximately 36 hours postoperative, the serum sodium had risen to 161 meq/L despite previous conversion to hypotonic saline. Inappropriate diuresis was suspected, possibly due to nephrogenic diabetes insipidus (NDI). Further evaluation consisted of urine electrolytes that revealed a sodium level of 27 meq/L and an osmolality of 167 mosm/kg. A serum sample was submitted for laboratory evaluation of AVP. Serum osmolality was noted to be 330 mosm/kg. Using 100kg for lean body weight, the free water deficit was calculated to be approximately 10.2 liters. Replacement using D5W was initiated to replace half this deficit over the ensuing 24 hours.
|
Because of high urine volume of low specific gravity and a clear inability to concentrate urine in the face of hypertonic dehydration, urine output replacement was initiated cc for cc with D2.5 in 1/8 NS (Fig. 1). This solution was selected to minimize dextrose administration and avoid possible glycosuria. Over the ensuing 5 postoperative days, urine output reached a maximum of 1200 cc/hour. Fluid administration of this magnitude with hypotonic solution required central access, which also allowed measurement of central venous pressure (CVP). An empiric trial of dDAVP (20 mic intranasal) was attempted without apparent effect on urine output or concentration, and was subsequently discontinued. Tube feedings were begun via the jejunostomy with an elemental diet and increased as tolerated to meet his estimated protein calorie requirements (Harris-Benedict equation) by postoperative Day 3. An early attempt to under-replace urine output resulted in a repeat episode of hypernatremia.
In discussing the patient's status with family members, it was discovered that he routinely consumes several quarts of ice water per day. In addition, he can consume a case (24 twelve oz. cans) of soda within an hour when it is available. He drinks because of thirst and does not ascribe any unusual or special powers to drinking water; such magical thinking is characteristic of psychogenic polydipsia. This history of polydipsia was first noted apparently 6 years before this hospitalization and worsened over the past 3 years with polyuria and nocturia. These symptoms were not reported to his physicians.
|
By postoperative Day 10, the patient was taking liquids PO and was restarted on hydrochlorothiazide. By postoperative Day 13 he was tolerating a regular diet, with stable electrolytes and serum osmolality and was subsequently discharged. He was not restarted on lithium.
The patient's Uosm reached a maximum of 285 mosm/kg. A serum AVP level was returned after discharge and noted to be 19 pcg/mL at a time when the serum sodium was 162 meq/L and the Uosm was 144 mosm/kg. This level of AVP is extraordinarily high. The low baseline Uosm and absent rise ( <50 per cent) of Uosm with dDAVP confirmed the diagnosis of nephrogenic diabetes insipidus.3 A diagnosis of primary or psychogenic polydipsia is unlikely to produce a concentrating defect of this magnitude unless the patient drinks 16 to 20 liters/day. In this case, the patient was unable to drink perioperatively.
The mechanism by which lithium is thought to cause nephrogenic diabetes insipidus centers upon a failure of water reabsorption in the distal tubule. Cellular uptake of lithium is thought to impair the activity of an AVP-sensitive adenyl cyclase associated with cell membranes. This enzyme is responsible for generating intracellular cAMP, which acts as the second messenger for AVP and PTH. The inhibitory effect on tubular water reabsorption is thought to be mediated by this decrease in intracellular cAMP. The mechanism by which lithium inactivates adenyl cyclase activity and maintains enzymatic inactivation after lithium is withdrawn is still unknown.
Certain, often reversible, histological changes have been associated with lithium-induced nephrogenic diabetes insipidus: tubular cytoplasmic swelling, glycogen deposition, dilatation of tubules, and microcyst formation. Interestingly enough, the renal biopsy findings from one patient with persistent nephrogenic diabetes insipidus were consistent with chronic interstitial nephritis. Additionally, it has been suggested experimentally that chronic tubulo-interstitial changes may occur following long-term treatment with lithium.4 As demonstrated by this case presentation, lithium may occasionally induce severe and permanent defects in urinary concentration. The concentrating defect most often improves after cessation of lithium, though this often occurs over a prolonged period of time.
Although seemingly paradoxical, thiazide diuretics have proven beneficial in the management of NDI. This type of diuretic is known to inhibit sodium reabsorption in the cortical diluting segment of the nephron. This action is ultimately responsible for the antidiuretic effect in these patients. The obligate increase in urinary sodium excretion and losses result in contraction. Proximal tubular water and sodium reabsorption then will increase, resulting in less water being delivered to the collecting tubules. Although no net gain in concentration is garnered, a smaller urine volume is presented to the distal nephron, and subsequently a decrease in urine volume is noted. Thiazide diuretic induced hypovolemia can also improve urinary concentration by increasing medullary interstitial osmolality through unknown mechanisms.
Finally, specific nonsteroidal anti-inflammatory drugs have an indication in the treatment of NDI.5 The exact mechanism of antidiuresis attributed to these medications has not been entirely elucidated. There is some thought that decreased prostaglandin synthesis, perhaps by cyclo-oxygenase inhibition, plays a major role. Indomethacin appears to be the most efficacious. However, it has been demonstrated that other NSAIDS will provide varying or situational dependent anti-diuresis. It should be noted that, like the thiazide diuretics, these medications provide no increase in urine concentration but only a decrease in urine volume. Even so, the therapeutic ramifications are significant. Urine outputs of 10-12 liters/day can be reduced to more manageable volumes of 3 or 4 liters.
Nephrogenic diabetes insipidus can present a formidable clinical problem in the perioperative period. A high index of suspicion and timely evaluation can direct appropriate fluid replacement and medical therapy. The cornerstones of therapy are limited administration of osmolar loads, free access to water, and when that is not possible, careful monitoring of serum electrolytes (specifically sodium) and osmolality with free water replacement. With limitation of sodium in IV fluids and early aggressive return to GI tract water administration via the jejunostomy, the patient's Uosm rose, although never to greater than serum levels. Concurrently, his urine output decreased dramatically over a 4-5 day period. Other measures taken during that time were to replace the losses from the patient's NG with solute free fluid in order to further reduce his obligate water diuresis; and to reinstitute his prior HCTZ as early as possible postoperatively. Because his primary surgical pathology was a complication of peptide ulcer disease, he was not considered a candidate for indomethacin treatment.
Thus, vigilant attention not only to fluid administration but also to osmolar electrolyte balance is needed in order to effectively manage NDI in postoperative surgical patients who require emergent volume resuscitation in conjunction with oral fluid deprivation or restriction.
REFERENCES
- Boton R, Gaviria M, Baile DC. Prevalence, pathogenesis and treatment of renal dysfunction associated with chronic lithium therapy. AM J Kidney Dis 1987;10:329-45.
- Singer I, Forrest JW. Drug-induced states of nephrogenic diabetes insipidus. Kidney International 1976;10:82-95.
- Robertson G. Differential diagnosis of polyuria. Ann Rev Med 1988;39:425-42.
- Walker RG, Escott M, Birhcall J, et al. Chronic progressive renal lesions induced by lithium. Kidney Int 1986;29:875-81.
- Allen HM, Jackson RL, Winchester MD, Deck LV, Allon M. Indomethacin in the treatment of lithium-induced nephrogenic diabetes insipidus. Arch Int Med 1989;149:1123-6.
|
Departments of General Surgery and Medicine, Keesler Medical Center, Keesler Air Force Base, Mississippi (M.A.J., J.O., G.H.G., M.W.P.) November 1, 1994 Address correspondence and reprint requests to Michael W. Paluzzi, M.D., AETC 81st MG/SGHS, 301 Fisher St., Suite 204, Keesler AFB, MS 39534-2527. |
This translation by the NDI Foundation is to assist the lay reader. To provide a clear, accessible interpretation of the original article, we eliminated or simplified some technical detail and complicated scientific language. We concentrated our translation on those aspects of the article dealing directly with NDI. The NDI Foundation thanks the researchers for their work toward understanding and more effectively treating this disorder.
© Copyright NDI Foundation 2007 (JC)
Nephrogenic diabetes insipidus (NDI) is a well-known complication of long-term lithium therapy which occurs in 12% to 33% of patients treated with lithium. (It is not clear whether the authors are saying that 12%-33% of lithium patients develop NDI.) NDI is characterized by excessive, watery urine despite high arginine vasopressin (AVP) levels. (AVP tells the kidneys to retain water. When one urinates frequently and the urine is quite watery, the kidneys are not retaining water but are passing the water into urine. Sometimes they need to do this; the body needs to get rid of that water. But sometimes the body needs to hold onto the water, and the kidneys should retain it; AVP tells the kidneys to do that.) This disorder is tolerated because it is accompanied by extreme thirst, thus many patients get the fluid they need, but it can cause major, irreversible problems when oral fluid intake is curtailed as it is before and during surgery.
Our report involves a lithium patient with previously undiagnosed NDI who required anesthesia, surgery, and oral fluid deprivation to repair a perforated duodenal ulcer.
Case Report
Our patient is a 42-year-old man with a history of schizophrenia and mild hypertension who had been using lithium for about 12 years. He came to the emergency room because during the previous 24 hours he had had abdominal pain, nausea, no appetite, and he'd been vomiting. Radiographic evaluation revealed a moderate amount of subdiaphragmatic free air.
The patient was prepared for surgery; fluid resuscitation was initiated with normal saline. During the repair of his ulcer, a feeding jejunostomy was placed. (The jejunum is a part of the small intestine. A jejunostomy is the surgical creation of a permanent opening between the jejunum and the abdominal wall.)
After surgery major problems involving fluid resuscitation arose. The patient's increase in urine output became unusually high. This was initially thought to be caused by overly aggressive fluid resuscitation. After further evaluation and tests, NDI was suspected. The urine that was expelled was replaced with an equal amount of solution (to prevent dehydration). Over the next 5 days urine output reached 1200 cc per hour. Tube feedings were begun via the jejunostomy with an elemental diet and increased as tolerated. By the third day following his operation, his food had been increased to meet his protein calorie requirement. An early attempt to under-replace urine output resulted in an episode of hypernatremia (too much sodium in the blood).
Free water was gradually added to his feedings through the jejunostomy and simultaneously subtracted from the urine output replacement. By the 10th day following the operation he was drinking liquids, and by the 13th day he was tolerating a regular diet. Lithium was not recommenced.
Through discussions with family members we learned of important symptoms that he'd never revealed to his physicians. He had experienced polydipsia (excessive thirst) over the past 6 years. It had increased over the past 3 years and been accompanied with polyuria (excessive urination) and nocturia (disturbed sleep because of a need to urinate). During this time he routinely drank several quarts of ice water each day. In addition, he could drink a case (24 twelve oz. cans) of soda within an hour when it was available. He drank because he was thirsty and did not ascribe any unusual or special powers to drinking water (such magical thinking is characteristic of psychogenic polydipsia, which is feeling thirsty even though the body physiologically does not need liquid at that time).
After discharge a serum AVP level was checked and found to be very high. This, in part, confirmed the diagnosis of NDI.
Discussion
The mechanism by which lithium is thought to cause nephrogenic diabetes insipidus centers upon a failure of water reabsorption in the distal tubule (a part of the kidneys). Lithium impairs the activity of an AVP-sensitive enzyme. This enzyme generates intracellular cAMP which acts as the second messenger for AVP and PTH. AVP tells the kidneys to retain water rather than passing it into the urine. If AVP cannot do what it is supposed to do, the kidneys will not retain the water that the blood needs, and this will cause problems in the body. (It's not clear what PTH is.) This case demonstrates that lithium may impair the body's ability to concentrate urine. This inability may improve after lithium use is stopped, though it may take a long time.
Although seemingly paradoxical, thiazide diuretics (drugs which increase urine production) have proven beneficial in managing NDI. This type of drug inhibits sodium reabsorption and ultimately has an antidiuretic (decreasing the amount of water expelled in urine) effect in people. Though these drugs decrease the amount of urine expelled from the body, they do not increase the concentration of the urine. Thiazide diuretic induced hypovolemia (an abnormally low volume of blood circulating through the body) can also improve urinary concentration.
Specific nonsteroidal anti-inflammatory drugs (NSAID) have also been used in the treatment of NDI. Indomethacin appears to be the most effective. Other NSAIDs have also been used to control urine output. It should be noted that, like thiazide diuretics, these medications provide no increase in urine concentration, but only a decrease in urine volume. Urine volumes of 10-12 liters/day can be reduced to more manageable volumes of 3 or 4 liters.
NDI can present formidable problems for surgical patients. Extreme care must be taken to insure that patients receive the appropriate amounts of fluids.