Patterns of injury in mycophenolate mofetil-related colitis.

Mycophenolate mofetil (MMF) was introduced as a new immune-suppression drug in the mid-1990s. It is widely utilized in solid-organ transplantation immune-suppression regimens. Side effects include gastrointestinal (GI) toxicity in the form of nausea, vomiting, and diarrhea. Physicians tend to reduce the dose of MMF or switch their patients to an enterio-coated formula to overcome the side effects. Because GI side effects are well linked to MMF, colonoscopy is not utilized in most of the cases to investigate the diarrhea. However, Crohn's disease-like changes in the colon, erosive enterocolitis, and graft versus host disease-like colonic changes associated with the use of MMF have been reported. Colonic findings in five patients whose symptoms resolved after substituting another agent for MMF are described in the present report. Repeat colonoscopy 4 months following discontinuation of MMF showed reparative changes in one of our patients. MMF is an important drug in organ transplantation immune-suppression regimens; however, with its widespread usage, additional side effects continue to be recognized.

Cardiovascular effects of vasopressin following V(1) receptor blockade compared to effects of nitroglycerin.

Studies to more clearly determine the mechanisms associated with arginine vasopressin (AVP)-induced vasodilation were performed in normal subjects and in quadriplegic subjects with impaired efferent sympathetic responses. Studies to compare the effects of AVP with the hemodynamic effects of nitroglycerin, an agent that primarily affects venous capacitance vessels, were also performed in normal subjects. Incremental infusions of AVP following V(1)-receptor blockade resulted in equivalent reductions in systemic vascular resistance (SVRI) in normal and in quadriplegic subjects. However, there were major differences in the effect on mean arterial pressure (MAP), which was reduced in quadriplegic subjects but did not change in normal subjects. This difference in MAP can be attributed to a difference in the magnitude of increase in cardiac output (CI), which was twofold greater in normal than in quadriplegic subjects. These observations are consistent with AVP-induced vasodilation of arterial resistance vessels with reflex sympathetic enhancement of CI and are clearly different from the hemodynamic effects of nitroglycerin, i.e., reductions in MAP, CI, and indexes of cardiac preload, with only minor changes in SVRI.

Inducible nitric oxide synthase in the bladder of spinal cord injured patients with a chronic indwelling urinary catheter.

PURPOSE: Spinal cord injured patients are at increased risk for bladder carcinoma. Nitric oxide production in areas of chronic inflammation may provide a stimulus for carcinogenesis by serving as a source of nitrosating agents that generate potentially carcinogenic nitrosamines from secondary amines normally present in urine. MATERIALS AND METHODS: To determine whether inducible nitric oxide synthase is expressed as a catalyst for sustained nitric oxide production by cellular elements in chronically inflamed bladder mucosa immunohistochemical studies were performed on mucosal biopsies obtained from 37 adults with spinal cord injury. All participants had required a chronic indwelling urethral or suprapubic catheter for greater than 8 years. RESULTS: Histopathological studies revealed active inflammatory infiltrates in all 37 biopsy specimens, squamous metaplasia in 20, epithelial dysplasia in 3 and carcinoma in 1. Inducible nitric oxide synthase was detected in inflammatory cells localized to the lamina propria. Inducible nitric oxide synthase positive cells were identified as macrophages using monoclonal antibodies to macrophage antigen. There was no inducible nitric oxide synthase expression in the urothelial cell layers. Immunostaining for inducible nitric oxide synthase was not detected in bladder mucosal biopsy specimens obtained from cadaveric organ donors. CONCLUSIONS: Inducible nitric oxide synthase is expressed in inflammatory macrophages in areas of chronic inflammation in the bladder mucosa of spinal cord injured patients with a chronic indwelling bladder catheter. The expression of inducible nitric oxide synthase may potentially lead to the sustained production of nitric oxide and its oxidative products, the nitrosation of urinary amines and the formation of potentially carcinogenic nitrosamines in the bladder.

Acute fulminant lactic acidosis complicating metastatic cholangiocarcinoma.

A patient with cholangiocarcinoma, metastatic to the liver and lungs, developed acute fulminant lactic acidosis in the absence of overt hepatic failure, sepsis, hypoxia, or circulatory failure. Despite extensive tumor replacement of hepatic parenchyma, no acid-base disorder was present during initial evaluation. The onset of acute lactic acidosis was temporally associated with the development of otherwise asymptomatic episodes of intermittent atrial arrhythmias. Once established, lactic acidosis was inexorably progressive, despite resolution of arrhythmias. Extensive areas of acute necrosis within the large hepatic metastases were demonstrated on postmortem examination, suggesting that local tissue ischemia, precipitated by cardiac arrhythmias, lead to excessive lactic acid production.

Risk factors for development of proteinuria in chronic spinal cord injury.

A retrospective, case-control study was performed to investigate the risk factors that may contribute to the development of proteinuria in patients with chronic spinal cord injury (SCI). During an 18-month period, 31 subjects with a 24-hour protein excretion of 1.0 g or greater were identified. Three control subjects with SCIs with a 24-hour urinary protein excretion of less than 1.0 g during the same time period were randomly selected for each study subject with proteinuria. Clinical data, including level and duration of injury, age, presence of indwelling bladder catheter, number of decubitus ulcer procedures, serum albumin and creatinine concentrations, hematocrit, creatinine clearance, and the presence of hypertension and diabetes mellitus, were obtained from medical records. Subjects with proteinuria had other evidence of renal dysfunction with greater serum creatinine concentrations and reduced creatinine clearances, serum albumin concentrations, and hematocrits. Proteinuric subjects were older, had a longer duration of injury, had undergone a greater number of decubitus ulcer procedures, and were more likely to have hypertension and indwelling bladder catheters. The independent predictors for the development of proteinuria using logistic stepwise multiple linear regression analysis were the use of chronic indwelling bladder catheters, number of decubitis ulcer procedures, presence of hypertension, and older age. These data suggest that inflammatory complications associated with complications of chronic SCI, rather than SCI per se, contribute to the development of proteinuria. SCI patients with proteinuria have more impaired renal function and increased mortality compared with SCI patients without proteinuria.

Hemodynamic response to vasopressin in dehydrated human subjects.

BACKGROUND: Despite the known potent vasoconstrictor effects of vasopressin, the role of this hormone in the maintenance of blood pressure is incompletely understood. In studies performed in animals with increased plasma vasopressin concentrations, several complex cardiovascular effects have been noted, including decreases in heart rate and cardiac output, which may account for a lack of effect on arterial pressure despite the vasopressin-induced increase in total peripheral resistance. Only a few studies have been done to assess the cardiovascular effects of vasopressin in human subjects, and most of these have been limited to measurement of heart rate and arterial pressure only. The present study was designed to identify more fully the cardiovascular effects of vasopressin when plasma vasopressin concentrations are increased by osmotic stimulation without the superimposition of major nonosmotic stimuli associated with severe volume depletion. METHOD: Studies were performed on 11 normal human subjects in supine and erect posture before and after 24 hours of fluid deprivation, and following administration of a selective V1 receptor antagonist, [d(CH2)5Tyr(ME)]AVP, after dehydration. Cardiovascular parameters were measured noninvasively by thoracic electrical bioimpedance cardiography and blood samples for measurements of plasma concentrations of vasopressin and other hormones affected by dehydration and differences in posture were collected for subsequent analysis. RESULTS: After 24 hours of fluid restriction, plasma osmolality was increased from 287 +/- 0.9 to 294 +/- 0.7 mosm/kg H20 and plasma vasopressin concentrations (Pavp) were increased in both supine and erect posture. Mean arterial (MAP) and systolic blood pressure (SBP) were reduced by fluid restriction but were higher in erect than in supine posture both before and after fluid restriction. Heart rate (HR), diastolic blood pressure (DBP), and systemic vascular resistance (SVRI) were also higher in erect than in supine posture, while cardiac index (CI), stroke index (SI), end-diastolic index (EDI), and an index of total thoracic fluid content (TFC) were all reduced in erect posture, both before and after dehydration. Plasma renin activity (PRA) and plasma norepinephrine concentrations (Pne) were increased in erect posture, both before and after dehydration, but there was no effect of erect posture on plasma vasopressin concentrations (Pavp), either before or after dehydration. Administration of the V1 receptor antagonist after dehydration had no effect on hemodynamic parameters other than small reductions in DBP and cardiac preload. CONCLUSION: It is concluded from these studies that small increases in Pavp associated with moderate dehydration do not play a role in the maintenance of arterial pressure in normal human subjects in either supine or erect posture.

Inappropriate antidiuresis and hyponatremia with suppressible vasopressin in Guillain-Barré syndrome.

Studies in which plasma osmolality was altered acutely by oral water loading and hypertonic sodium chloride infusion were performed to further identify the mechanisms involved in the pathogenesis of the syndrome of inappropriate secretion of antidiuretic hormone (SIADH) in a patient with Guillain-Barré syndrome. Although resetting of the osmotic threshold for vasopressin release was demonstrated in these studies, this does not seem to have been a primary factor in the development of SIADH in this patient. Downward resetting of the osmotic threshold by sustained hypoosmolality has been previously demonstrated, and it is possible that this may account for the initially low osmotic threshold identified by our studies. These studies suggest that inappropriate antidiuresis, as shown by the absence of a diuretic response to low threshold suppression of the plasma arginine vasopressin concentration was due either to a vasopressin-independent mechanism or to markedly increased renal tubular sensitivity to vasopressin.

Effects of vasopressin V1-receptor blockade during acute and sustained hypovolemic hypotension.

The response of vasopressin and its role in the maintenance of arterial pressure during and after development of hypotensive central hypovolemia were studied in tilt table studies in quadriplegic subjects. The studies were performed during acute head-up tilting to a maximally tolerated degree of tilt (8 subjects) and during sustained head-up tilt following a 20% reduction in mean arterial pressure (MAP) (11 subjects). Studies in all subjects were performed on two separate days, once with and once without administration of a selective vasopressin V1-receptor antagonist. During acute head-up tilting, plasma vasopressin concentrations (PAVP) did not increase significantly until MAP decreased to approximately 60 mmHg at maximal tilt. There was no difference in the degree of hypotension produced in the presence compared with the absence of V1-receptor blockade. There was also no difference in plasma renin activity (PRA) or in plasma cortisol or aldosterone concentrations at maximal tilt. In contrast, during sustained head-up tilt following a 20% reduction in arterial pressures, systolic and mean arterial pressures were significantly lower and PRA was significantly higher in the presence than in the absence of V1-receptor blockade. PAVP increased and was significantly higher after 30 min of sustained tilt than pretilt PAVP in supine posture. These studies do not provide evidence of a role for vasopressin in the maintenance of arterial pressure during the acute development of hypotensive hypovolemia in human subjects, but they do provide evidence of a modest role for vasopressin in the maintenance of arterial pressure when the effect of hypovolemia is more moderate and sustained.

Hemodynamic response to vasopressin during V1-receptor antagonism in baroreflex-deficient subjects.

Six quadriplegic subjects and 6 control subjects received high-dose arginine vasopressin (AVP) infusions at rates of 500, 1,000, 2,000, and 4,000 microU.kg-1.min-1 in consecutive 10-min intervals. Six additional quadriplegic subjects received low-dose AVP infusions at rates of 50, 100, 200, 400, and 800 microU.kg-1.min-1. All subjects were studied once with and once without administration of a selective V1-receptor antagonist. During high-dose AVP infusions without V1-receptor blockade, mean arterial pressure (MAP) increased from 80 +/- 4 to 87 +/- 5 mmHg (P < 0.05) in quadriplegic subjects but was unchanged in control subjects. In the presence of V1-receptor blockade, MAP decreased from 75 +/- 4 to 58 +/- 4 mmHg (P < 0.001), and heart rate (HR) increased from 61 +/- 5 to 80 +/- 5 beats/min (P < 0.001) in quadriplegic subjects. In the studies on control subjects, MAP decreased only from 75 +/- 3 to 72 +/- 5 mmHg (P < 0.05), whereas HR increased from 64 +/- 4 to 87 +/- 4 beats/min (P < 0.001). Plasma renin activity (PRA) increased in both quadriplegic and control subjects. The effects of low-dose AVP infusions on MAP, HR, and PRA in quadriplegic subjects were similar to those observed during high-dose infusions. Thus, in the absence of baroreceptor-mediated sympathetic nervous system responses, a vasodilatory effect of AVP that is capable of producing marked reductions in MAP can be demonstrated in the presence of V1-receptor blockade.

Characteristics of vasopressin release during controlled reduction in arterial pressure.

Because of the interruption of the descending sympathetic nervous pathways, individuals with cervical spinal cord injury experience orthostatic hypotension when in an upright posture. The changes in hemodynamic parameters that occur during upright posture can be closely monitored and quantitated during progressive head-up tilting on a tilt table. We have utilized this method to assess the response of vasopressin and other vasoactive hormones to gradual, progressive reductions in arterial pressure and to identify possible threshold responses to baroreceptor stimulation in human subjects. Studies were performed in 12 quadriplegic subjects, 3 paraplegic subjects, and 3 normal control subjects. Data from the studies in paraplegic and normal subjects did not differ and were pooled as control data. In quadriplegic subjects, mean arterial pressure (MAP) decreased from 93 +/- 4 mm Hg to 60 +/- 3 mm Hg in a closely correlated (r = 0.948, p < 0.002) linear relationship with increasing degrees of tilt, whereas in control subjects, MAP increased from 81 +/- 4 to 88 +/- 3 mm Hg. Plasma vasopressin concentrations (Pavp) increased minimally in quadriplegic subjects until MAP was reduced to levels that were 25% to 30% lower than MAP with subjects in the supine posture. Beyond this level of hypotension, Pavp increased markedly. Log-linear regression analysis of these data showed a highly significant correlation (r = 0.85, p < 0.0002) between in Pavp and MAP, which defines Pavp as an exponential function of decreasing MAP. Changes in Pavp in control subjects were minimal during incremental head-up tilting. In contrast, plasma renin activity (PRA) increased in both quadriplegic and control subjects. Log-linear regression analysis of these data showed highly significant correlates between in PRA and degree of tilt in both quadriplegic (r = 0.958, p < 0.0002) and control (r = 0.873, p < 0.0002) subjects. Plasma atrial natriuretic peptide concentrations decreased linearly with increasing degrees of tilt. The rate of decline in Panp was greater in quadriplegic than in control subjects. These studies provide additional evidence that Pavp increases exponentially as a function of decreasing MAP and suggest that a critical threshold level of hypotension exists at which vasopressin release accelerates rapidly in response to baroreceptor stimulation. At this level of reduced MAP, Pavp reaches levels that are potentially capable of exerting a pressor effect.

Case report: acute renal vein thrombosis in patients with spinal cord injury and secondary amyloidosis.

Chronic spinal cord injury, when complicated by chronic suppurative infections, has replaced chronic tuberculosis as a leading cause of secondary amyloidosis. Renal involvement with secondary amyloidosis is characterized by the presence of nephrotic range proteinuria and an increased incidence of renal vein thrombosis. Two cases of acute renal vein thrombosis associated with secondary amyloidosis in patients with spinal cord injury are presented. In both cases, a past history of extensive decubitus ulcerations and urinary tract infections preceded the development of nephrotic range proteinuria. In case 1, nonoliguric acute renal failure occurred after the development of acute bilateral renal vein thrombosis. The patient declined dialytic therapy and expired with uremia. In case 2, worsening renal function and increased proteinuria resulted after the development of acute unilateral renal vein thrombosis. These cases include the clinical and anatomic findings of acute renal vein thrombosis that occur as a complication of secondary amyloidosis. Acute renal vein thrombosis should be considered whenever an acute change in renal function or increase in proteinuria is noted in this setting.

Vasopressin-independent alterations in renal water excretion in quadriplegia.

Postural effects on water excretion are known to be increased in patients with cervical spinal cord injury and may result in marked impairment of the ability to excrete a water load, especially in erect posture. Both vasopressin-dependent and vasopressin-independent mechanisms have been implicated. To assess the roles of these mechanisms and further identify the factors involved in the renal response to erect posture, sustained water loading studies were performed on 11 quadriplegic subjects and 9 healthy control subjects, supine and erect (sitting). Renal blood flow was assessed by p-aminohippurate clearance (CPAH) measurements in 7 quadriplegic and 5 control subjects. During maximal water diuresis, plasma vasopressin concentrations were reduced to unquantifiable levels in all subjects. Osmolar clearance, free water clearance (CH2O), and distal delivery of filtrate (DDF) were all lower in quadriplegic than in control subjects, supine and erect. The relationship between CH2O and DDF was the same in quadriplegic as in control subjects and was not altered by change in posture in either group. Creatinine clearance and CPAH were lower in erect than in supine posture in quadriplegic subjects but not in control subjects. We conclude that impairment of water excretion in stable normonatremic quadriplegic subjects can be attributed primarily to vasopressin-independent mechanisms involving reduced filtrate delivery to diluting segments of the renal tubules rather than to resistance to normal suppression of vasopressin release.

Reversible vasopressin deficiency in severe hypernatremia.

Studies to assess the relationship between plasma arginine vasopressin concentration (Pavp) and plasma osmolality (Posm) were performed on an elderly patient with dementia who developed severe hypernatremia due to inadequate water intake following a debilitating hip fracture. Serum sodium concentrations were 174 and 196 mEq/L on two consecutive hospital admissions. During the second of these admissions, sequential measurements of Pavp and Posm were obtained as hypernatremia was gradually corrected. Pavp during this period was correlated with Posm (r = 0.887, P < 0.01), but was low despite the presence of hyperosmolality and volume depletion. Pavp decreased from 0.56 microU/mL to 0.18 microU/mL as Posm decreased from 396 to 338 mOsm/kg H2O. The regression line of this relationship intercepted the abscissa at 320 mOsm/kg H2O. Hypertonic sodium chloride infusion to reassess this relationship 2 days following the correction of hypernatremia increased Pavp only to 0.67 microU/mL while increasing Posm from 297 to 316 mOsm/kg H2O. Nevertheless, Pavp and Posm were significantly correlated (r = 0.937, P < 0.001). The slope of the regression line was 0.031, and Posm at the abscissal intercept was 292 mOsm/kg H2O. A similar increase in Posm from 290 to 310 mOsm/kg H2O during hypertonic sodium chloride infusion 11 days following the correction of hypernatremia increased Pavp to 1.95 microU/mL (r = 0.786, P < 0.05). The magnitude of the increase in Pavp at this time was equivalent to that previously observed in studies of normal subjects. The slope (0.048) and abscissal intercept (280 mOsm/kg H2O) of linear regression were also consistent with observations in studies of normal subjects.(ABSTRACT TRUNCATED AT 250 WORDS)

Coexistence of anti-neutrophil cytoplasmic antibody-associated glomerulonephritis and membranous glomerulopathy.

The association of renal Wegener's granulomatosis with other glomerular diseases is very rare. A case of anti-neutrophil cytoplasmic antibody-associated necrotizing glomerulonephritis superimposed on a membranous glomerulopathy in a patient with systemic Wegener's granulomatosis is reported. Renal failure was corrected by immunosuppressive therapy treatment, but a non-nephrotic-range proteinuria persisted for several months. The association of membranous glomerulopathy with anti-glomerular basement membrane disease and other autoimmune diseases is well described; however, anti-neutrophil cytoplasmic antibody-associated vasculitis superimposed on membranous glomerulopathy has not been reported previously.

Reversible renal failure due to IgA nephropathy associated with osteomyelitis.

Clinical features of acute glomerulonephritis, with microscopic hematuria, red blood cell (RBC) casts, proteinuria, and acute renal insufficiency developed in a patient with chronic osteomyelitis. Before the development of osteomyelitis, renal function and findings on urinalysis were normal. Complete eradication of osteomyelitis by surgical amputation led to resolution of the abnormal urinary findings, and renal function returned to near pre-osteomyelitis levels. Although acute glomerular disease has been reported to occur as a rare complication of osteomyelitis, the unique feature of the present case was the histological finding of IgA nephropathy. There was no arthritis, purpura, skin rash, or gastrointestinal involvement to suggest a diagnosis of Henoch-Schönlein purpura and there was no evidence of chronic liver disease. The temporal relationship between the onset of the renal disease, which followed the development of chronic osteomyelitis, and its resolution following removal of the focus of infection, suggests that the IgA nephropathy may have been related directly to the osteomyelitis (secondary IgA nephropathy). Glomerular diseases associated with chronic bacterial infections, including osteomyelitis, are discussed, with emphasis on infections that have been associated with the development or exacerbation of IgA nephropathy.

Chronic hyponatremia due to resetting of the osmostat in a patient with gastric carcinoma.

A 54-year-old schizophrenic patient who presented with hyponatremia and nephrotic-range proteinuria was subsequently discovered to have a gastric adenocarcinoma. Psychogenic water drinking, sodium depletion, and cardiac, adrenal, hepatic, and thyroid disease were excluded as causes of hyponatremia. The serum creatinine concentration was normal, and, although renal biopsy showed changes consistent with immune complex glomerulopathy, proteinuria remitted without treatment. Moderately severe hyponatremia persisted, and the diagnosis of gastric adenocarcinoma was made after the onset of early satiety 1 year later. Surgical exploration at the time of partial gastric resection revealed local metastatic lymph node involvement. Following the patient's uneventful recovery from surgery, studies of osmoregulation of vasopressin release and renal water handling were performed to determine the cause of chronic hyponatremia refractory to sodium chloride administration. Oral water loading studies revealed normal urinary diluting ability and appropriate suppression of plasma vasopressin concentrations. However, hypertonic sodium chloride infusion studies revealed a highly significant correlation between plasma osmolality and plasma vasopressin concentration, and a low osmotic threshold for vasopressin release based on linear regression analysis of the plasma vasopressin response to increasing plasma osmolality. Low osmotic threshold for vasopressin release was confirmed by exponential (log linear) and parabolic methods of data analysis. The findings in these studies are consistent with the typical features of the reset osmostat variant of the syndrome of inappropriate antidiuresis. To our knowledge, this is the first report of the occurrence of this syndrome in association with gastric adenocarcinoma.

Modulation of plasma aldosterone by physiological changes in hydrogen ion concentration.

To assess the effect of extracellular hydrogen ion concentration (PH+) on aldosterone secretion, studies in which other known modulators could be controlled were performed on 13 patients undergoing hemodialysis. High (35 mM) or low (14-17 mM) dialysate bicarbonate concentrations were utilized on separate days to either decrease or increase PH+, while plasma potassium concentrations (PK) were held at constant levels and changes in plasma renin activity (PRA) were minimized by avoiding changes in body weight. Changes in PH+ were associated with concordant changes in plasma aldosterone concentration (Pa) in both high- and low-bicarbonate studies. When these changes in Pa in high- and low-bicarbonate studies were analyzed together as a function of corresponding changes in PH+, a significant correlation could be demonstrated (r = 0.659, P less than 0.001). There was no correlation between changes in Pa and changes in PK, plasma sodium, plasma adrenocorticotropic hormone (ACTH), or PRA. Using the same methods to control PH+ and other variables during hemodialysis, the effects of altered PH+ on ACTH-stimulated aldosterone and cortisol secretion were evaluated in studies on six patients who received incremental infusions of ACTH after pretreatment with dexamethasone. In these studies, there was no demonstrable effect of PH+ on Pa or plasma cortisol concentration. We conclude that physiological changes in PH+ have a weak modulating effect on basal aldosterone secretion that may not be evident in the presence of other acutely applied stimuli.

Immunoglobulin A nephropathy in a renal allograft of a black transplant recipient.

Although IgA nephropathy (IgA N) is reportedly the most common form of primary glomerulonephritis worldwide, there is a very low incidence of IgA N in Black Americans, and IgA N in Black patients with renal allografts has not been reported. In this report, we present the case of a Black patient with endstage renal disease presumed secondary to hypertensive nephrosclerosis who developed nephrotic range proteinuria due to IgA N in a cadaveric renal allograft 2 years following transplantation. Biopsy of the allograft in the immediate post-transplantation period had revealed no evidence of IgA N. Chronic active hepatitis related to hepatitis C preceded the development of proteinuria by approximately 1 year, raising the possibility that IgA N in the renal allograft was secondary to the liver disease. The clinical and histological features of primary IgA N and IgA N secondary to liver diseases are discussed.

Reversible changes in osmoregulation of vasopressin release due to impaired water excretion.

Studies of renal water handling and the effects of altered hydration and posture on the osmoregulation of vasopressin release were performed on a chronically hyponatremic patient with complete cervical spinal cord transection at the C-5 level. Acute oral water loading studies showed marked reduction in free water clearance and urine diluting ability, despite appropriate suppression of plasma vasopressin concentrations. Orthostatic reductions in arterial blood pressure during head-up tilting and following the assumption of sitting posture were also demonstrable, and may have contributed to, but could not fully account for, the defect in renal water excretion, which persisted in supine posture. Hypertonic sodium chloride infusion studies performed before fluid restriction showed that low preinfusion plasma osmolality was associated with a reduced osmotic threshold for vasopressin release, which was subsequently corrected by a period of fluid restriction that restored the patient's plasma osmolality to a normal level. This shift in osmotic threshold can be inferred from both linear regression and log-linear regression analysis of the data. These studies show that marked impairment of renal water excretion coupled with unrestricted water intake can result in altered osmoregulation of vasopressin release in association with persistent plasma hypo-osmolality, which can be corrected by fluid restriction.