Isotonic Hyponatremia Following Transurethid Prostate Resection David M. Rothenberg, MD,* Arnold S. Berns, MD,? Anthony D. Ivankovich, MD$ Department

of Anesthesiology,

ical Center; and Department Reese Hospital and Medical

The proper treatment of hyponatremia during transurethral resection of the prostate continues to be controversial. Two cases of isotonic hyponatremia are reported here, and the literature regarding the incidence and treatment of hyponatremia during transurethral resection of the prostate is reviewed. In each case, the patient developed neurologic changes during complicated transurethral prostate resection. Despite the rapid decrease in the serum sodium concentration, serum osmolalityremained normal due to the resorption of the bladder irrigant glycine. Therefore, etiologies other than cerebral edema are postulated as the cause of the neurologic mani-

Rush-Presbyterian-St.

Luke’s

of Medicine, Renal Division, Center, Chicago, IL.

MedMichael

festations. Also, the role of the osmolar gap in directing appropriate therapy is emphasized in an effort to avoid unnecessary use of hypertonic saline. Finally, an appropriate differential diagnosis of the neurologic changes seen during the transurethral resection of the prostate syndrome is discussed. Keywords: Transurethral resection of prostate; complications; hyponatremia; isotonic; osmolar gap.

Introduction

*Assistant Professor, Department of Anesthesiology, Rush Medical College TAssistant Clinical Professor of Medicine, Pritzker School of Medicine SProfessor and Chairman, Medical College

Department

of Anesthesiology,

Rush

Address reprint requests to Dr. Rothenberg at the Department of Anesthesiology, Rush-Presbyterian-St. Luke’s Medical Center, Jelke 739, 1753 West Congress Parkway, Chicago, IL 60612, USA. Received for publication March 7, 1989; cepted for publication June 27, 1989. 0 1990 Butterworth

48

revised manuscript

Publishers

J. Clin. Anesth.,

vol. 2, Jan/Feb

1990

ac-

Acute hyponatremia following intravascular absorption of irrigating solutions is a well-described complication of transurethral resection of the prostate. With the use of osmotically active irrigating solutions, the concomitant intravascular absorption of osmotically active solute occurs, and marked shifts in brain water are thereby attenuated. Indeed, since distilled water has been replaced by these agents, major neurologic sequelae and death are rarely seen due to this complication. Assessment of serum osmolality and calculation of the osmolar gap in differentiating acute isotonic hyponatremia from the potentially more critical acute hypotonic hyponatremia in the setting of transurethral resection of the prostate are described in the nephrology and urology literature. These concepts,

Isotonic hyponatremia and prostate surgery: Rothenberg et al.

however, are not as well described in the anesthesia literature, and therefore they are often overlooked by the anesthesiologists who routinely treat these patients. In an effort to provide a better understanding of the pathophysiology of this disorder, two cases of isotonic hyponatremia as a complication of transurethral resection of the prostate are reported here, and the literature is reviewed. The usefulness of the osmolar gap in correctly diagnosing and treating these patients is emphasized.

Case Reports Case 1 A 66-year-old male (72 kg) with hypertension and benign prostatic hypertrophy was admitted with symptoms of urinary tract obstruction. Past medical history was otherwise unremarkable. The patient’s medications included hydralazine 50 mg, propranolol 40 mg, and nifedipine 10 mg, all taken three times a day, and one tablet hydrochlorothiazide/triamterene taken daily. The physical examination was remarkable only for normal blood pressure (BP) and an enlarged prostate. Admitting laboratory data included the following: serum sodium, 143 meq/L; serum potassium, 4.7 meq/L; serum chloride, 106 meq/L; serum bicarbonate, 25 meq/L; BUN, 59 mg/dl; serum creatinine, 3.3 mg/dl; serum glucose, 95 mg/dl; hemoglobin, 12 gm/dl; electrocardiogram (EKG), left ventricular hypertrophy; chest radiograph, normal. Three days after relief of the urinary tract obstruction and improvement of the serum creatinine to 1.5 mg/dl, a transurethral resection of the prostate was performed. Preoperatively, the patient received all antihypertensive medications except hydrochlorothiazide/triamterene. Baseline vital signs included BP, 135/85 mmHg, and heart rate (HR) 68 beats/minute. A subarachnoid block was performed with a 25-gauge spinal needle at the L4-5 level, with 12 mg of tetraCaine administered after cerebrospinal fluid (CSF) aspiration. A sensory level to approximately T6 was achieved. Vital signs remained stable. Saline 0.9% was used as maintenance intravenous (IV) fluid therapy. No sedation was administered. Surgery commenced with the use of 1.5% glycine as a bladder irrigant. Approximately 30 minutes after the onset of surgery, a prostatic venous sinus was perforated. At the time, a total of approximately six 3liter bags of 1.5% glycine had been used as bladder irrigation, and 300 ml of 0.9% saline had been administered as IV fluid. Hypotension and bradycardia with ST segment depression on EKG ensued, and the

patient became confused. Atropine, ephedrine, and 100% oxygen were administered, with stabilization of both BP and HR. The surgery was discontinued, and the patient was transported to the recovery room. Initial laboratory values in the recovery room were as follows: serum sodium, 116 meq/L; serum potassium, 4.2 meq/L; serum glucose, 100 mg/dl; BUN, 55 mg/dl; calculated serum osmolality, 256 mOsm/kg * H,O (defined as 2 x serum sodium + glucose/l8 + BUN12.8); measured serum osmolality, 298 mOsm/ kg * H,O; osmolar gap, 42 mOsm/kg - H,O (defined as measured osmolality minus calculated osmolality). Despite the knowledge of the serum osmolality and the osmolar gap, therapy was initiated with IV furosemide and 3% saline. The follow-up serum sodium concentration after 90 minutes was 126 meq/L. The serum sodium on postoperative day 2 had returned to baseline. No neurologic or cardiovascular sequelae were noted, and the patient was discharged on the sixth postoperative day.

Case 2 A 75-year-old male (68 kg) with a history of benign prostatic hypertrophy presented with symptoms of urinary tract obstruction. He had hypertension treated with metoprolol 50 mg daily. The physical exam on admission was remarkable for an enlarged prostate and slightly distended bladder. Admitting laboratory data included the following: serum sodium, 14 1 meq/ L; serum potassium, 4.4 meq/L; serum chloride, 98 meq/L; serum bicarbonate, 26 meq/L; BUN, 29 mg/ dl; serum creatinine, 2.4 mg/dl; serum glucose, 100 mg/dl; hemoglobin, 15.2 gm/dl; EKG, normal; chest radiograph, normal. An indwelling bladder catheter was inserted, with relief of the urinary obstruction, and a transurethral resection of the prostate was performed. The patient received metoprolol as a preoperative medication. Baseline vital signs included BP, 165188 mmHg; and HR, 62 beats/minute. A subarachnoid block was performed with a 25-gauge spinal needle at the L4-5 level, with 12 mg of tetracaine administered after CSF aspiration. A sensory level of T, was achieved, and vital signs remained stable. Midazolam 2 mg was given as intraoperative sedation. Saline 0.9% was administered as maintenance IV fluid. Surgery commenced with the use of 1.5% glycine as a bladder irrigant. Approximately 90 minutes after the onset of surgery, the prostatic capsule was perforated. At the time, approximately nine 3-liter bags of 1.5% glycine had been used as bladder irrigation, and 750 ml of 0.9% saline had been administered as IV fluid. Shortly thereafter, the patient became conJ. Clin. Anesth.,

vol. 2, JanlFeb

1990

49

Case Reports

fused, and labored respiration and hypertension were noted. The EKG showed new T wave inversions. Oxygen 100% was administered by a face mask, and the patient was intubated via direct laryngoscopy. Surgery was discontinued. IV furosemide 5 mg was administered, and after stable vital signs were achieved, the patient was transported to the recovery room. A chest radiograph showed pulmonary edema. The EKG showed nonspecific ST-T wave changes. Laboratory values in the recovery room were as follows: serum sodium, 102 meq/L; serum potassium, 5.3 meq/L; serum glucose, 105 mg/dl; BUN, 21 mg/ dl; calculated serum osmolality, 2 17 mOsm/kg 1 H,O; measured serum osmolality, 286 mOsm/kg * H,O; osmolar gap, 69 mOsm/kg * H,O. Therapy continued with IV furosemide and fluid restriction. Follow-up serum sodium values at 3 and 6 hours postoperatively were 125 meq/L and 134 meq/ L, respectively. The signs and symptoms of pulmonary edema resolved over the next 24 hours, and the patient was extubated without further cardiovascular or neurologic sequelae. The patient was discharged on the seventh postoperative day.

Discussion Transurethral resection of the prostate gland was first described in the late 1920s. Extravasation of irrigating solution was first recognized by Rudnik’ in 1935, with major morbidity and mortality secondary to this complication reported in 1947.* Landsteiner and Finch” described the pathophysiology, attributing the complication of shock to massive hemolysis secondary to absorption of hypotonic irrigating solution (i.e., distilled water). Subsequently, the introduction of nonhemolytic irrigating solutions has essentially eliminated the problem of intravascular hemolysis during transurethral resection of the prostate. For the most part, however, these solutions are hypotonic, and despite eliminating hemolysis, intravascular absorption of them may lead to acute and profound hyponatremia. Major neurologic sequelae during transurethral resection of the prostate have been felt to be due to this rapid decrease in serum sodium. In 1956, Harrison et al4 described a patient who became hypertensive and apprehensive during the course of a prolonged transurethral resection of the prostate (using glycine irrigating solution and under spinal anesthesia), then suddenly complained of restlessness, nausea, and blindness. The patient vomited and was noted to be cyanotic, subsequently developing vascular collapse. Serum sodium is noted to have decreased from 140 to 111 meq/L. These authors4 con50

J. Clin. Anesth.,

vol. 2, Jan/Feb

1990

eluded that “the urological patient who rapidly absorbs irrigating fluid is analogous to the patient with diarrhea who drinks large volumes of water but does not replace electrolytes lost. An acute, sudden increase in blood volume with an electrolyte-poor substance is well known to precipitate water intoxication, the signs and symptoms of which have been extensively discussed in the literature. We believe that the concomitant increase in cerebral blood volume and cerebral edema causes an increase in intracranial pressure. Cerebral edema is probably secondary to osmolar fluid shifts from the hypotonic extracellular fluid compartment to the relatively hypertonic intracellular spaces.” The phenomenon of acute hyponatremia leading to cerebral edema during transurethral resection of the prostate continues to be reported. As recently as 1980, Henderson and Middleton retrospectively attributed 14 cases of coma after transurethral resection of the prostate (using glycine irrigating solution) to the patient’s acute hyponatremic state. They concluded that “the transurethral resection reaction syndrome and post-transurethral resection of the prostate coma are the result of cerebral edema secondary to osmolar fluid shifts from the hypotonic extracellular spaces. Hyponatremia [after transurethral resection of the prostate] is truly a hypo-osmolar state since sodium makes up a large part of the serum osmolality.” In 1985, Rhymer and coworkers” prospectively measured serum sodium in 100 patients undergoing transurethral resection of the prostate in an attempt to correlate the transurethral resection of the prostate syndrome (defined as bradycardia, hypotension, and a confusional state in association with profound hyponatremia) with the use of postoperative glycine bladder irrigation. Of note was a 7% frequency of postoperative decrement in serum sodium (6 to 32 meq/L change). All seven patients were felt to manifest the syndrome as described, presumably secondary to the acute decrease in serum sodium. One patient expired on the second postoperative day. Despite an acute decrease in serum sodium from 135 meq/L to 103 meq/L, postmortem examination failed to show any evidence of cerebral edema. Unfortunately, these author+ failed to report serum osmolality in their studies. Desmond’ described 72 patients in whom he simultaneously measured serum osmolality and serum sodium concentration before and after transurethral resection of the prostate using 1.2% glycine irrigating solution. Of the patients studied, 53% developed postoperative hyponatremia, 28% of whom had a greater than 10 meq/L decline in serum sodium. Of these 38

Isotonic hyponatremia and prostate surgery: Rothenberg et al.

patients, only two manifested signs and symptoms compatible with cerebral edema. Both had concomitant decreases in serum osmolality greater than 30 mOsm/kg * H,O. He concluded that significant amounts of osmotically active glycine entered the vascular compartment, rendering serum osmolality near normal and thereby protecting the majority of the patients against cerebral edema. In the case of excessively rapid absorption or poor myocardial function, however, pulmonary and cerebral edema, secondary to volume overload and acute dilutional hyponatremia, may ensue. Desmond stressed that “because glycine maintains normal serum osmolality, we have not found it necessaary to give sodium. . . . [AIdministration of sodium (i.e., hypertonic saline) to a patient with normal osmolality might be dangerous because his plasma will become hyperosmolar.” In 1979, KirschenbaumS reported a similar scenario in a patient who developed acute hyponatremia posttransurethral resection of the prostate, with the serum sodium decreasing from 133 to 99 meq/L. A simultaneous serum osmolality was normal, and the patient manifested no signs or symptoms of cerebral or pulmonary edema. Kirschenbaum also emphasized the importance of differentiating acute hypotonic hyponatremia from acute isotonic hyponatremia, concluding that “aggressive treatment of the hyponatremia with hypertonic saline solutions in an individual with a normal serum osmolality could result in the development of severe hyperosmolality of the extracellular fluid in an otherwise benign condition.” Obviously, there is still a debate in the literature regarding the frequency, pathophysiology, and treatment of the transurethral resection of the prostate syndrome. An increased likelihood of the transurethral resection of the prostate syndrome appears to be associated with large tissue resection, prolonged surgery, and excessive height of the irrigating solution relative to the symphysis pubis, leading to an increase in hydrostatic prostatic fossa pressure.g The c!inical manifestations of irritability, confusion, visual changes, nausea, vomiting, seizures, coma, and, occasionally, death, are well documented.*s5J0 Attributing this constellation of symptoms solely to acute hyponatremiainduced cerebral edema, however, does not appear to be fully warranted. The key to interpreting these neurologic changes is in differentiating hypotonic hyponatremia from isotonic hyponatremia, as well as recognizing that other potential etiologies of this syndrome exist, including acute volume overload resulting in pulmonary edema and subsequent hypoxemia, as well as hyperammonemic encephalopathy secondary to the metabolism of excess glycine or hyperglycinemia itself.11-14

In assessing the nature of the hyponatremia in the transurethral resection of the prostate syndrome, an understanding of the interplay among serum sodium, osmolality, and tonicity is required. The normal physiologic range of serum osmolality is approximately 280 to 290 mOsm/kg * H,O, with the normal determinants of serum osmolality being sodium, glucose, and urea. As previously stated, serum osmolality may be measured directly or calculated. Normally, these two values agree within 5 to 10 mOsm/kg * H,O. When the measured serum osmolality exceeds the calculated osmolality by greater than 10 mOsm/kg * H,O, an increased osmolar gap exists.15 This fact implies that unmeasured solute must be present in plasma water in significant amounts and therefore may influence net movement of water across cells. Tonicity, or effective osmolality, is ordinarily a function of serum sodium. Since urea permeates cells freely, it does not infuence cellular water shifts. In contradistinction, substances such as glycine, mannitol, and sorbitol, all agents used as irrigating solutions during transurethral resection of the prostate, are relatively impermeable. Accumulation of these solutes in the plasma, as seen during prostatic sinus perforation, will indeed influence tonicity and net transcellular water movement. The brain itself responds like an osmometer, with intracellular swelling or shrinking dependent on decreased or increased effective plasma osmolality. Experimental data in animals suggest that an osmotic gradient between plasma and brain of at least 25 to 30 mOsm/kg * H,O is required to promote acute intracellular edema.16,17 Therefore, provided that a relatively normal effective osmolality is maintained, cerebral edema should not occur. In both patients reported in this study, as well as in others cited in the literature referenced above, an acute decrease in the serum sodium concentration occurred secondary to intravascular absorption of significant amounts of hypotonic irrigating solution. However, a large amount of osmotically active glycine also was absorbed simultaneously, generating a marked osmolar gap. Serum osmolality remained near normal despite severe hyponatremia, and the neurologic changes that occurred may have been due to hypoxemia, ammonia toxicity, glycine toxicity, or possibly cation-induced alterations in glial transmembrane potential,18Jg but it is very unlikely that they were due to acute cerebral edema (Table 1). Unfortunately, the relationship between the serum sodium concentration and serum osmolality and the use of the osmolar gap were overlooked in the first case presented here. These patients represent a classification of hyponatremia best characterized as isotonic hyponatremia. J. Clin. Anesth., vol. 2, Jan/Feb

1990

51

Case Reports Table 1. Differential

Diagnosis

of Altered

Neurologic

Status

during

Transurethral

Resection

of the Prostate

Normal Serum Osmolality, Elevated Osmolar Gap Volume Hypoxemip

overload

with pulmonary

edema

< Excess sedation

with inadequate

due to excess glycine metabolism

Encephalopathy

Hyperammonemia

Hyperglycinemia