Hyponatremia following transurethral resection, hysteroscopy, or other procedures involving electrolyte-free irrigation

INTRODUCTION — A variety of transurethral, hysteroscopic, and other percutaneous procedures utilize large volumes of irrigating/distension solutions. These procedures include transurethral resection of the prostate (TURP), transurethral resection of bladder tumors, hysteroscopic transcervical diagnostic and therapeutic procedures (eg, resection of submucosal leiomyomas), percutaneous removal of kidney stones, and some other percutaneous and minimally invasive procedures.

The electrosurgery devices that have been used in past years during these procedures have a monopolar design and cannot be used with electrolyte-containing irrigation fluids (eg, isotonic saline or Lactated Ringer). Several different nonconductive (ie, nonelectrolyte) solutions have been employed. The use of these nonelectrolyte solutions has been associated with a number of complications including hyponatremia, a variable degree of hypoosmolality, and certain solution-specific problems described below.

However, bipolar electrosurgery devices and various laser and microwave systems are now much more widely used and are as effective as monopolar devices [1]. These newer devices are compatible with electrolyte-containing irrigation and distension solutions [2-6]. Use of isotonic saline or Ringer's lactate solution will not generate hyponatremia. However, absorption of a large volume of saline solution will expand the extracellular fluid volume and may generate volume overload with dyspnea, pulmonary edema, hyper- or hypotension, and hyperchloremia [7].

Hyponatremia following use of nonconductive (ie, nonelectrolyte) irrigation solutions for transurethral resection of the prostate or bladder or for hysteroscopy will be reviewed here. General principles of diagnosis and management of hyponatremia and of TUR and hysteroscopy are discussed separately:

●(See "Diagnostic evaluation of adults with hyponatremia".)

●(See "Overview of the treatment of hyponatremia in adults".)

●(See "Surgical treatment of benign prostatic hyperplasia (BPH)".)

●(See "Clinical presentation, diagnosis, and staging of bladder cancer".)

●(See "Overview of hysteroscopy".)

●(See "Hysteroscopy: Managing fluid and gas distending media", section on 'Fluid media'.)

NONCONDUCTIVE IRRIGATION FLUIDS — Three major nonconductive (ie, nonelectrolyte) fluids are used for surgical procedures:

●1.5 percent glycine (most commonly used)

●3 percent sorbitol

●5 percent mannitol

Glycine and sorbitol solutions are hypoosmotic, having an osmolality of 200 and 165 mosmol/kg, respectively, compared to the normal serum osmolality of 280 to 296 mosmol/kg. By contrast, the 5 percent mannitol solution is almost isosmotic to plasma (275 mosmol/kg).

INCIDENCE — Hyponatremia following transurethral resection, hysteroscopy, or other percutaneous procedures using nonconductive (ie, nonelectrolyte) irrigation solutions is an uncommon event. The severity of hyponatremia is directly related to the volume of irrigation fluid that is retained. Patients with a decrease in serum sodium of 10 mEq/L or more are likely to develop neurologic symptoms. The volume of irrigant fluid absorption required to develop this degree of hyponatremia is smaller in females than in males. (See 'Volume and rate of fluid absorption' below and 'Pathogenesis of neurologic symptoms' below.)

Based upon data from large multicenter series, hyponatremia severe enough to produce confusion and nausea and vomiting (each of which can also be due to other factors) occurs in approximately 2 percent of males undergoing TUR of the prostate (this is often called the "post-TURP syndrome") [4,8], while fluid overload and/or hyponatremia occurs in 0.06 to 0.2 percent of females undergoing hysteroscopy [9,10]. Hyponatremia can also occur after percutaneous nephrolithotomy [11].

RISK FACTORS — The major risk factor for hyponatremia following transurethral resection, hysteroscopy, or other percutaneous procedures utilizing nonelectrolyte solutions is the volume of irrigant absorbed. Absorption of 1000 mL of glycine irrigation fluid in females undergoing hysteroscopy is associated with a reduction in serum sodium of approximately 10 mEq/L (figure 1), while 3000 mL reduces the serum sodium by approximately 30 mEq/L [12-14].

Males undergoing TUR of the prostate show smaller reductions of 6 to 8 mEq/L after absorbing 1000 mL and 20 mEq/L after absorbing 3000 mL [14-16]. These sex differences are probably due mostly to higher lean body weight and total body water in males. Thus, males must absorb more fluid than females to produce the same dilutional reduction in serum sodium. (See 'Halt procedure at absorption thresholds' below.)

Procedural risk factors for excess fluid absorption and hyponatremia during endoscopic procedures include the type of procedure (operative versus diagnostic), introduction of fluid at high pressure, visceral perforation, prolonged operative time, and type of anesthesia [17].

Operative procedures are associated with a higher risk of excess fluid absorption than diagnostic procedures. When tissue is resected, veins may be severed (eg, large prostatic veins) or uterine venous sinuses may be exposed [14,17]. Irrigation fluid can then be directly and rapidly absorbed into the vascular system. In hysteroscopy, for example, the risk of excess fluid absorption is particularly high during resection of fibroids. The risk is much lower with diagnostic procedures since monopolar electrosurgery is not performed and nonconductive (ie, nonelectrolyte) fluids are not required.

Fluid absorption into the vascular space begins when the fluid pressure exceeds the venous pressure (approximately 10 mmHg) [17]. (See 'Minimize fluid pressure' below.)

In addition to fluid absorption directly into the vascular space (intravasation), perforation of a viscus during surgery can result in rapid extravasation of irrigation fluid into the surrounding space. With perforation of the bladder or uterine wall, fluid will extravasate into the peritoneal cavity or retroperitoneum. With perforation of the prostatic capsule during transurethral resection (TUR), fluid extravasates into the periprostatic space [14,17-19]. Fluid can also traverse the Fallopian tubes during hysteroscopy; however, a history of prior tubal ligation does not seem to affect the volume of fluid absorbed [17,20]. The fall in serum sodium occurs more slowly following extravasation compared to intravasation. (See 'Time course of hyponatremia' below.)

Prolonged operative time permits greater opportunity for fluid absorption [21-23]. However, the available data are not sufficient to define an operative time threshold beyond which excessive fluid absorption is likely to occur. For TUR of the prostate, fluid absorption is greater during the second half of the procedure, regardless of operating time [23]. (See 'Monitor fluid absorbed' below.)

The type of anesthesia may influence fluid absorption in females undergoing endometrial resection, although the available data are conflicting. In a randomized trial of 24 females undergoing hysteroscopy, fluid absorption was significantly higher with epidural compared to general anesthesia (648 versus 381 mL) [24]. By contrast, a subsequent retrospective cohort study found that fluid absorption was highest with general anesthesia, less with epidural anesthesia, and lowest with local anesthesia and intravenous sedation [25].

The administration of intracervical vasopressin to reduce bleeding during hysteroscopy appears to protect against the development of hyponatremia by reducing local absorption of the irrigation fluid. In a randomized trial of 33 females undergoing hysteroscopic endometrial ablation, intracervical administration of vasopressin during hysteroscopy was associated with a significantly lower plasma glycine concentration at 20 minutes (8.8 versus 16 mmol/L) and a nonsignificantly smaller reduction in plasma sodium from baseline (2.2 versus 6.8 mEq/L at 40 minutes) [26]. Possible protection against hyponatremia was also noted in an observational study of intraprostatic administration of vasopressin in males undergoing TUR of the prostate [27]. In contrast to the possible benefits of intracervical or intraprostatic vasopressin, the systemic effects of vasopressin would be expected to exacerbate any reduction in serum sodium by limiting water excretion.

There are limited data regarding patient-specific risk factors for excess fluid absorption and hyponatremia following TUR of the prostate or bladder or hysteroscopy. For males undergoing TUR of the prostate, smoking was associated with fluid absorption exceeding 1000 mL in a retrospective study [28]. The physiologic basis of such an effect, if real, might be that the blood supply to the prostate increases due to the variable degree of hypoxemia that accompanies smoking.

PATHOGENESIS OF HYPONATREMIA

Volume and rate of fluid absorption — The likelihood of developing hyponatremia and associated symptoms and complications (eg, nausea, cerebral edema) following transurethral resection (TUR) of the prostate or bladder or hysteroscopy is related to the rate, volume of absorption, osmolality, and volume of distribution of the nonconductive (ie, nonelectrolyte) irrigation solution, and the rapidity of metabolism and/or excretion of the solute and the water load [12,13,17,29-34]. There is a direct correlation between the glycine deficit (the difference between the infused and collected volumes of the glycine irrigant) and the fall in serum sodium (figure 1). (See 'Clinical manifestations' below.)

The relationship between the glycine deficit and the reduction in serum sodium and the development of nausea and cerebral edema on computed tomography was illustrated in a study of 20 females [13]:

●Ten females who absorbed less than 500 mL of glycine irrigant had a mean decrease in serum sodium of 2.5 mEq/L and did not have symptoms of hyponatremia or changes suggestive of cerebral edema on computed tomography.

●Six females who absorbed 1000 mL or more had a reduction in serum sodium of 13 to 25 mEq/L; all had nausea and changes suggestive of cerebral edema on computed tomography.

●Four females absorbed 500 to 1000 mL of glycine irrigant; three had cerebral edema, two of whom had reductions in serum sodium of 10 and 13 mEq/L.

As noted above, a higher volume of irrigant fluid absorption is required to produce the same degree of hyponatremia in males who have a higher lean body weight and total body water than females. (See 'Risk factors' above.)

The rate of fluid absorption is probably also an important factor in the development of hyponatremia. TUR and hysteroscopy procedures generally last for one hour or less, so fluid absorption is relatively rapid. One study suggested that a rate of fluid absorption in excess of 200 to 300 mL per 10 minutes is required to develop hyponatremia [30].

Fluid absorption thresholds at which the procedure should be halted are discussed below. (See 'Halt procedure at absorption thresholds' below.)

Time course of hyponatremia — The time course of the development of hyponatremia varies with the site of fluid absorption. Fluid absorbed into the vascular space (intravasation) rapidly lowers the serum sodium. This fluid presumably enters the vascular space directly through vessels opened during the procedure.

The time course is more prolonged when perforation of a viscus during surgery leads to extravasation of irrigation fluid into the surrounding space. The serum sodium is usually lowest one to two hours after surgery [15] and the degree of hyponatremia is only approximately one-third that expected from absorption directly into the vascular space [15,35]. (See 'Risk factors' above.)

Subsequent events depend in part upon the irrigant used:

●Glycine enters cells over several hours and, by four hours, glycine is almost equally distributed between the extracellular and intracellular compartments [33]. Glycine entry into cells is associated with osmotic water movement from the extracellular fluid into the cells, which is probably responsible for the gradual increase in serum sodium that begins as soon as 10 to 20 minutes after the development of hyponatremia [13,30,33,36].

In a series of 20 patients described above, the six patients with a glycine deficit (the difference between the infused and collected volumes of the glycine irrigant) of 1000 mL or more and two patients with a glycine deficit of approximately 700 mL had a decrease in serum sodium of 10 mEq/L or more [13]. The mean fall in serum sodium was 16 mEq/L (141 to 125 mEq/L) immediately after surgery. The serum sodium then increased by a mean of 8.4 mEq/L in the first four hours after surgery, was stable for the next four hours, and then continued to increase toward baseline.

Glycine is metabolized to serine, glyoxylate, ammonia, and other products and approximately 5 to 10 percent is excreted unchanged in the urine.

●Sorbitol is metabolized to glucose and fructose in the liver and then to carbon dioxide and water. Approximately 5 to 10 percent is excreted unchanged in the urine.

●Mannitol does not enter cells and is not metabolized; it is excreted entirely in the urine.

The urinary excretion of mannitol and, to a much lesser degree, glycine and sorbitol creates an osmotic diuresis. The associated electrolyte-free water loss can raise the serum sodium concentration and the increase in sodium excretion will enhance correction of the volume overload [37].

Time course of osmolar shifts — The serum or plasma osmolality is normally determined by the sum of the osmotic contributions of the three major solutes in the plasma: sodium salts, glucose, and urea. The osmotic contributions of glucose and BUN to plasma osmolality are usually small. Thus, in the absence of diabetes mellitus or kidney failure, the serum sodium concentration is the most important determinant of serum osmolality.

These contributions are illustrated in the following formula, which can be used to estimate the serum osmolality (S-osm):

(Note glucose and BUN are measured in units of mg/100 mL.)

The serum sodium is multiplied by two to account for the accompanying anion (chloride and bicarbonate) and the divisors 18 and 2.8 convert units of mg/dL into mosmol/L. (See "Serum osmolal gap".)

As with the serum sodium concentration, the serum osmolality varies over time following TUR or hysteroscopy. The hyponatremia associated with irrigation fluids is associated with a fall in serum osmolality that is less pronounced than with other etiologies of hyponatremia (eg, volume depletion, syndrome of inappropriate antidiuretic hormone secretion) or, in some cases, absent, due to presence of glycine, sorbitol, or mannitol in the extracellular fluid.

Glycine (200 mosmol/kg) and sorbitol (165 mosmol/kg) solutions are hypoosmotic compared to the normal serum osmolality of 280 to 296 mosmol/kg. Thus, the initial serum osmolality will be reduced with systemic infusion or absorption of glycine or sorbitol-containing irrigants. By comparison, the 5 percent mannitol solution is near isosmotic and the initial serum osmolality will be relatively unchanged.

With all three irrigation fluids, there will be a large serum (or plasma) osmolal gap between the measured serum osmolality, which will include the infused organic solute, and the serum osmolality estimated from the above formula, which does not include the irrigant solute. (See "Serum osmolal gap".)

These relationships have been described during both TUR of the prostate [36,38-42] and hysteroscopy [32,33,43,44]. In a prospective study of 100 patients who underwent TUR with glycine irrigant, for example, 10 patients developed neurologic symptoms [42]. The mean serum sodium fell by an average of 17.7 mEq/L, which should have lowered the serum osmolality by approximately 35 mosmol/kg. However, the serum osmolality fell by only 11.4 mosmol/kg due to the osmotic effect of glycine. The modest reduction in serum osmolality alone would not be expected to produce such marked neurologic symptoms. (See 'Pathogenesis of neurologic symptoms' below.)

The serum osmolal gap gradually disappears due to glycine entering cells, the metabolism of glycine and sorbitol, and urinary excretion of glycine, sorbitol, and mannitol. As a result, the serum osmolality will gradually fall toward the level calculated from the serum sodium, glucose, and urea concentrations [33,45].

One might expect the urine to be maximally dilute (less than 100 mosmol/kg) at this time to permit excretion of the excess water. However, most patients also have postoperative, stress-related release of antidiuretic hormone (ie, syndrome of inappropriate ADH secretion), which impairs water excretion and slows correction of the hyponatremia [46]. Other factors that can impair urinary dilution are the associated osmotic diuresis, which, as mentioned above, is most pronounced and lasts for a longer time with mannitol irrigation, and possibly direct stimulation of antidiuretic hormone release by glycine [47,48]. (See 'Time course of hyponatremia' above and "Pathophysiology and etiology of the syndrome of inappropriate antidiuretic hormone secretion (SIADH)", section on 'Surgery'.)

PREVENTION — Fluid overload and hyponatremia can be prevented by the following general measures:

●Avoid the use of nonconductive (ie, nonelectrolyte) irrigation fluids, if possible

●Minimize fluid absorption by using low infusion pressures and limiting the duration of surgery

●Monitor the quantity of fluid absorbed

●Stop the procedure at predetermined absorption thresholds

Avoid use of nonconductive irrigants — When monopolar electrosurgery is used, nonconductive irrigation fluids are required to allow electrical energy to be focused and thereby avoid thermal burns. By contrast, bipolar systems permit the use of electrolyte solutions such as isotonic saline. Isotonic electrolyte fluids virtually eliminate the risk of procedure-related hyponatremia [2-6,49-52]. In a randomized trial of hysteroscopy in 200 postmenopausal females, for example, there was no reduction in serum sodium in the bipolar electrode group that used isotonic saline irrigant compared with a mean 5 mEq/L reduction in serum sodium in the monopolar electrode group that used glycine irrigant [50].

A meta-analysis comparing monopolar versus bipolar TURP found that the bipolar technique eliminated the "TURP syndrome" and had a lower overall complication rate [4]. The mean fall in sodium concentration was 5.1 mEq/L with unipolar TURP and 1.5 mEq/L with bipolar TURP, but this difference was not statistically significant. A subsequent international multicenter double-blind randomized controlled trial not included in this meta-analysis compared the perioperative efficacy and safety of bipolar versus monopolar techniques for TURP [5]. Of the 295 patients enrolled, 279 underwent TURP, including 138 with monopolar and 141 with bipolar devices. In the bipolar group, the average fall in serum sodium was 0.8 mEq/L, and the lowest serum sodium recorded in any patient was 131 mEq/L. By comparison, in the monopolar group, the average fall in serum sodium was 2.5 mEq/L, and the lowest serum sodium recorded in any patient was 106 mEq/L. Perioperative efficacy, safety, and secondary outcomes were otherwise comparable in the two groups. Other randomized trials have reported similar results [6].

Other energy sources, such as laser and microwave, also permit the use of isotonic electrolyte-containing irrigation fluid. Although these fluids do not induce hyponatremia, their absorption may produce volume overload and hyperchloremia [7].

Monitor fluid absorbed — Several methods can be used to monitor the amount of fluid absorbed during surgery so that patients at risk of severe hyponatremia can be detected. A direct approach is calculation of the fluid deficit (also called the "glycine deficit"), which is the difference between the volume of irrigation fluid administered and the volume removed by suction. Fluid is collected through suction tubing attached to the operative device (hysteroscope or cystoscope) and in plastic drapes beneath the patient.

In hysteroscopy, the fluid deficit can be determined with an automated system or manually by the operating room staff. The automated system permits continuous measurement of the fluid deficit and intrauterine fluid pressure, titrates flow, and generates automatic alerts. An automated fluid pump and monitoring system is recommended by both the American College of Obstetricians and Gynecologists and the American Association of Gynecologic Laparoscopists [43,44]. There are no similar guidelines for urologic procedures.

With manual techniques, fluid is infused using the force of gravity (ie, elevating the infusion bag to different heights) or by inflating a large blood pressure cuff around the infusion bag. However, the effect of progressive elevation of the infusion bag on fluid absorption is not clear. In a study of 550 patients undergoing TUR of the prostate, there was no difference in fluid absorption within the range of 60 to 100 cm above the operating table [53].

A member of the surgical staff should be designated to measure the input and outflow and report the fluid deficit to the surgeon at regular intervals (eg, every 15 minutes). The accuracy of the volumetric method is reduced by two factors:

●Frequent overfilling of the infusion bags by the manufacturer by 3 to 6 percent [54].

●During TUR of the prostate, the accuracy of volumetric balance is diminished because of a higher degree of admixture with blood in the irrigating fluid returns, more frequent spillage on the floor, and urinary excretion [55].

Alternative techniques for monitoring fluid absorption include gravimetric weighing of the patient [56] and the respiratory output of tracer quantities of either ethanol or nitrous oxide that have been added to the irrigation fluid, a technique that is not commonly used in clinical practice [55,57-59].

Halt procedure at absorption thresholds — As noted above, serious complications of absorption of nonconductive irrigation fluids occur in most females who absorb more than 1000 mL, with a higher absorption threshold for males. (See 'Volume and rate of fluid absorption' above.)

To minimize the risk of hyponatremia, we suggest the following fluid absorption thresholds for females undergoing hysteroscopy using nonconductive (ie, nonelectrolyte) irrigation solutions. These are in keeping with the 2005 American College of Gynecology technology assessment [43]:

●A 750 mL fluid deficit represents excessive fluid absorption. The surgical team should temporarily halt the procedure. The fluid inflow should be stopped, suction should remain on to assist with fluid removal, and a sample sent for urgent measurement of serum sodium (see 'Evaluation' below). Assessment of mental status should be performed when possible but may be limited in patients under anesthesia. Once this information is available, a decision can be made to resume surgery for as short a period as possible or to terminate the procedure.

●At a 1000 to 1500 mL fluid deficit (depending in part upon patient size), the surgical team should terminate the procedure as quickly as possible, and the serum sodium concentration and mental status should be closely monitored. Many gynecologic surgeons plan to finish the case at a fluid deficit of 1000 mL, and immediately terminate the procedure at a fluid deficit of 1500 mL.

Higher fluid absorption thresholds, 1000 mL to temporarily halt the procedure and 2000 mL to terminate the procedure, have been suggested for TUR of the prostate. This approach is based in part upon an analysis of 273 patients who received glycine irrigant [60]. Symptoms of TUR syndrome (eg, nausea, vomiting, confusion, and/or hypotension) increased significantly when fluid absorption exceeded 1000 mL. The average number of such symptoms was 2.3, 3.1, and 5.8 per procedure with fluid absorption between 1000 and 2000 mL, between 2000 mL and 3000 mL, and more than 3000 mL, respectively. Symptoms developed more often 30 to 60 min postoperatively than during the surgical procedure. (See 'Clinical manifestations' below.)

The higher thresholds for fluid absorption in males reflect at least in part the greater lean body weight and therefore total body water in males. As a result, more fluid must be absorbed in males to produce the same reduction in serum sodium. (See 'Risk factors' above.)

Patients who are treated with saline irrigant (as with bipolar electrosurgery) have much less risk for hyponatremia, but can develop heart failure. The 2005 American College of Gynecology technology assessment recommends that the surgical team terminate the procedure as quickly as possible and that irrigant infusion be discontinued if fluid absorption reaches 2500 mL [43].

Minimize fluid pressure — The movement of irrigating fluid into blood vessels (intravasation) begins when the fluid pressure exceeds the venous pressure (approximately 10 mmHg) [17]. However, transurethral resection and hysteroscopy procedures occur within hollow organs. As a result, the pressure threshold for fluid absorption is best measured as the pressure within the operative site (prostate, bladder, or uterus). When this threshold is reached, fluid is absorbed through open, low-pressure venous channels or perforations induced by resection and, at high pressures, into tissue compartments. Irrigant absorption begins at intrauterine pressures of 70 mmHg or greater [61,62] and bladder pressures greater than 15 to 25 mmHg [63-66].

Limit operation time — Minimizing the duration of surgery is another approach to limiting fluid absorption [21,22]. Although there are no definitive data that have established an operative time threshold beyond which excessive fluid is predictably absorbed, we suggest that after one hour of surgery, the patient's overall status, the volume of fluid absorbed, and the anticipated time to completion of surgery be reassessed.

CLINICAL MANIFESTATIONS — The earliest symptom of hyponatremia associated with glycine irrigation is nausea, which commonly occurs when the serum sodium concentration has fallen by 10 mEq/L or more. This typically requires fluid absorption of at least 1000 mL in females undergoing hysteroscopy (figure 1) and a higher value in males undergoing TUR of the prostate. (See 'Risk factors' above.)

More severe hyponatremia, which has been associated with glycine irrigant absorption rates as high as 2000 to 3000 mL [60], can be associated with confusion, disorientation, twitching, seizures, hypotension, and transient visual abnormalities, such as decreased visual acuity or blurred vision [40,67-69].

Sorbitol and mannitol irrigation fluids are used less frequently. Compared to glycine irrigation, symptoms induced by excessive fluid absorption are similar with sorbitol irrigation (except for the lack of visual abnormalities) [31] and significantly less prominent with mannitol irrigation (which is not metabolized, is near isotonic, and remains extracellular) [70].

Absorption of large volumes of irrigation fluid can also cause expansion of the extracellular fluid and pulmonary edema [71,72].

Pathogenesis of neurologic symptoms — When acute hyponatremia occurs as a result of other etiologies (such as primary polydipsia, use of ecstasy), it is associated with acute hypoosmolality corresponding to the degree of hyponatremia. This acute hypoosmolality is associated with acute cerebral edema. The cerebral edema is due to a water shift into the brain from the hypotonic extracellular fluid and is thought to play a central etiologic role in the neurologic symptoms. (See "Manifestations of hyponatremia and hypernatremia in adults".)

However, the pathogenesis of neurologic symptoms following the absorption of nonelectrolyte irrigation fluids is less well defined. As mentioned in the preceding section, the hyponatremia associated with nonconductive (ie, nonelectrolyte) irrigation fluids is associated with a less pronounced or no fall in serum osmolality due to the presence of glycine, sorbitol, or mannitol in the irrigation fluid and the extracellular fluid. (See 'Time course of osmolar shifts' above and "Hysteroscopy: Managing fluid and gas distending media".)

One explanation relates to the difference between osmolality and tonicity. To the extent that glycine penetrates brain cells, it is an osmotically "ineffective osmole," which moves into cells with water [40,73]. However, factors other than hypoosmolality and cerebral edema may also cause the neurologic symptoms that often develop when treatment with nonelectrolyte irrigation fluids generates marked hyponatremia. The very low serum sodium concentration itself, glycine toxicity, and the accumulation of ammonia, serine, and/or glyoxylate (all generated by the metabolism of glycine) may be responsible [38-40,74-76].

Studies in experimental animals and humans are consistent with the hypothesis that multiple factors contribute to the toxicity of glycine and that neither hypoosmolality nor cerebral edema is essential for the development of neurologic symptoms:

●In a rat model, the mortality associated with extreme glycine-induced hyponatremia (90 to 95 mEq/L) was not reduced when the serum osmolality was maintained with a mannitol infusion [74].

●In another rat model, isoosmolar or hypoosmolar glycine infusions produced minimal cerebral edema even after glycine had equilibrated across most cell membranes [36].

●Although cerebral edema has been reported in a series of females who developed nausea following exposure to glycine irrigant, the findings were subtle [13]. The diagnosis of cerebral edema was based upon careful acute and post recovery computed tomography scans, and the diagnosis of cerebral edema could not have been made in any patient without this comparison.

Glycine and some of its metabolites, such as ammonia, may also have neurotoxicity [76,77]. The postoperative mental status of two patients who became comatose following TURP correlated more closely with the serum ammonia concentration than the serum sodium concentration [76]. However, the majority of patients undergoing TUR have a minimal or no increase in blood ammonia despite substantial absorption of glycine irrigant [78].

Glycine itself is an inhibitory transmitter in the central nervous system, especially the spinal cord, brainstem, and retina [79]. Glycine may be responsible for self-limited visual abnormalities including decreased visual acuity, blurred vision, and blindness [67-69].

Glycine is associated with more severe symptoms and a poorer outcome compared with alternative irrigating solutions, such as mannitol and sorbitol [80]. The metabolic and neurologic effects of glycine mentioned above probably account for this difference.

EVALUATION — Hyponatremia should be suspected whenever large volumes of nonconductive (ie, nonelectrolyte) irrigation fluids have been used, when operative time is prolonged (both of which promote excessive fluid absorption), or when patients develop new neurologic symptoms during postoperative recovery. The risk of neurologic symptoms is greatest when fluid absorption exceeds 1000 mL. In such patients, the serum sodium and osmolality should be urgently measured. (See 'Volume and rate of fluid absorption' above.)

The difference between the measured and calculated osmolalities (osmolal gap) reflects the concentration of the infused/absorbed irrigant and its nonionized metabolites. The calculated serum osmolarity is estimated from the following formula:

The serum osmolal gap, which is normally less than 5 to 10 mosmol/kg, can exceed 30 to 60 mosmol/kg immediately following transurethral resection or hysteroscopy due to the transient accumulation of glycine, sorbitol, or mannitol [33,39]. (See "Serum osmolal gap".)

MANAGEMENT — The management of hyponatremia following TUR, hysteroscopy, and some other procedures such as percutaneous nephrolithotomy is directed by the patient's clinical condition.

Asymptomatic hyponatremia — Specific therapy is not necessary for asymptomatic hyponatremia or, when symptoms cannot be assessed because of anesthesia, for a serum sodium reduction of 5 mEq/L or less (see 'Clinical manifestations' above). If kidney function is adequate, then the excretion of the excess water and metabolism and excretion of the infused nonelectrolyte solute will rapidly correct the hyponatremia. A loop diuretic can be given if the patient develops pulmonary congestion.

Symptomatic hyponatremia — Optimal therapy of patients with symptomatic hyponatremia varies with the serum sodium concentration, measured serum osmolality, and the volume status.

Role of hypertonic saline — Hypertonic saline is indicated in symptomatic patients with marked hyponatremia who have a substantially reduced serum osmolality or cerebral edema [40,41,43]. In addition to reducing cerebral edema, hypertonic saline replaces the potentially substantial urinary sodium losses that result from the osmotic diuresis caused by glycine, mannitol, or sorbitol. Kinetic modeling of fluid and electrolyte shifts in the "post-TURP syndrome" suggest that this osmotic-induced natriuresis both prolongs the period of hyponatremia and contributes to the hypovolemia and hypotension that often develops 15 to 20 minutes after absorption of the irrigant solution has ended [81].

The potential value of hypertonic saline in hyponatremic patients with substantial hypoosmolality was illustrated in a retrospective study of 18 patients who underwent TUR or hysteroscopy and developed symptomatic hyponatremia [41]. The mean serum sodium was 106 mEq/L and the mean serum osmolality was 235 mosmol/kg, with a serum glycine of 18 mmol/L (equivalent to 18 mosmol/kg), a value significantly lower than reported in other cases of glycine-induced hyponatremia of this severity (see 'Time course of osmolar shifts' above). Fourteen patients were treated with hypertonic saline and survived; neither the saline dose nor the rate of correction was reported. The other four patients were not treated with hypertonic saline and had a respiratory arrest and died even though the hyponatremia was less severe than in the survivors.

When hypertonic saline is given, we prefer the regimen that is recommended for the treatment of acute hyponatremia. The patient is initially given 100 mL of 3 percent saline as a bolus, which provides 51 mEq of sodium and should acutely raise the serum sodium concentration by 2 to 3 mEq/L. If the neurologic symptoms do not improve, a 100 mL bolus of 3 percent saline can be given one or two more times at 10-minute intervals [82]. (See "Overview of the treatment of hyponatremia in adults", section on 'Acute hyponatremia: Initial therapy (first six hours)'.)

Rapid correction of hyponatremia and hypoosmolality is probably safe following TUR or hysteroscopy. Because of the extremely short duration of hyponatremia, there has not been time for the cerebral adaptations that put patients with chronic hyponatremia at risk for osmotic demyelination if the serum sodium is raised too quickly. Nevertheless, the American College of Obstetricians and Gynecologists guidelines concluded that the serum sodium should not be raised by more than 12 mEq/L in the first 24 hours [43]. The rate of correction should be slower (8 to 10 mEq/L maximum increase in the first 24 hours) in patients who present with hyponatremia more than 48 hours after the procedure. (See "Manifestations of hyponatremia and hypernatremia in adults", section on 'Osmotic demyelination' and "Overview of the treatment of hyponatremia in adults".)

By contrast, optimal therapy has not been defined in symptomatic patients who have severe hyponatremia but a normal or near normal serum osmolality (eg, greater than 270 mosmol/kg) [40]. Hypertonic saline should probably not be used in such patients and hemodialysis may be the safest approach [40].

Role of hemodialysis — Hemodialysis will rapidly correct hyponatremia, osmotic derangements, volume expansion, and also remove the nonelectrolyte solute and its toxic metabolites (glycine, sorbitol or mannitol). It has been used in symptomatic patients with severe kidney disease [39] and in patients with severe neurologic symptoms, marked hyponatremia, and normal or only slightly reduced serum osmolality (which will be associated with a high serum osmolal gap) [40].

Loop diuretics — Absorption of large volumes of irrigation fluid can cause expansion of the extracellular fluid and pulmonary edema in patients undergoing transurethral resection or hysteroscopy [71,72]. Loop diuretics (eg, furosemide) can be given to treat fluid overload. However, loop diuretics should not be given in the absence of volume overload since they can reduce the effective intravascular volume and, as shown in two small randomized trials of patients undergoing TUR of the prostate, may contribute to the development of hyponatremia [83,84].

Possible role of surgical drainage — As discussed, perforation of a viscus during surgery can result in rapid extravasation of irrigation fluid into the surrounding space, and absorption of this fluid into the circulation can generate and contribute to persistence of, hyponatremia. (See 'Risk factors' above.)

Drainage of this fluid from the perivesical or peritoneal space might help treat the hyponatremia [85]. However, in a small series (eight patients), suprapubic drainage did not improve the hyponatremia and was associated with prolongation of hospitalization [15].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Fluid and electrolyte disorders in adults" and "Society guideline links: Hysteroscopy".)

SUMMARY AND RECOMMENDATIONS

●Transurethral resection (TUR) of the prostate or bladder, a variety of hysteroscopy procedures, and certain other percutaneous procedures (such as percutaneous nephrolithotomy) may use large volume irrigation with nonconductive (ie, nonelectrolyte) fluids to avoid thermal burns. Absorption of these irrigation fluids produces hyponatremia with varying degrees of hypoosmolality. Neurologic manifestations can occur when fluid absorption exceeds 500 mL and particularly 1000 mL. (See 'Incidence' above.)

These problems can be avoided with bipolar electrosurgery, which permits the use of isotonic saline irrigation. (See 'Avoid use of nonconductive irrigants' above.)

●Irrigant-associated hyponatremia can be prevented or minimized by reducing fluid absorption. This is accomplished with the use of low infusion pressures, limitation of surgery duration, by monitoring the degree of fluid absorption, and by aborting the procedure at prespecified absorption thresholds. (See 'Monitor fluid absorbed' above.)

●If a nonconductive solution is used, we suggest the use of an automated rather than a manual fluid monitoring system and terminating the procedure if the fluid deficit reaches 1000 to 1500 mL in females (depending in part upon patient size) and 2000 mL in males. (See 'Halt procedure at absorption thresholds' above.)

●Symptoms of hyponatremia include nausea and malaise. This may be followed by headache, lethargy, and obtundation, and eventually kidney failure, seizures, coma and respiratory arrest. Patients exposed to glycine irrigant may complain of visual deficits or blindness. (See 'Clinical manifestations' above.)

●Evaluation of patients who become symptomatic during or following one of these procedures, as well as the evaluation of all patients who have had prolonged surgery or a large volume of fluid absorbed includes measurement of the plasma sodium and osmolality and calculation of the plasma osmolal gap. (See 'Evaluation' above.)

●Specific therapy for modest hyponatremia is unnecessary in asymptomatic patients and when symptoms cannot be assessed because of anesthesia. (See 'Asymptomatic hyponatremia' above.)

●Patients with hypoosmolar hyponatremia who have symptoms of hyponatremia or who have severe hyponatremia should usually be treated with hypertonic saline. When hypertonic saline is given, we prefer the regimen that is recommended for the treatment of acute hyponatremia. The efficacy and safety of such therapy is uncertain in hyponatremic patients who have a normal or near normal serum osmolality (eg, greater than 275 mosmol/kg). (See 'Role of hypertonic saline' above and "Overview of the treatment of hyponatremia in adults", section on 'Acute hyponatremia: Initial therapy (first six hours)'.)

●There is little concern about overly rapid correction of the hyponatremia following TUR or hysteroscopy due to the extremely short duration of the hyponatremia. However, hyponatremia should not be overcorrected. (See 'Symptomatic hyponatremia' above.)

●Hemodialysis will rapidly correct hyponatremia, osmotic derangements, volume expansion, and also remove the nonelectrolyte solute and its toxic metabolites (glycine, sorbitol, or mannitol). It may be used in symptomatic patients with severe kidney disease. (See 'Role of hemodialysis' above.)

●For patients in whom major fluid absorption occurs following intraoperative perforation of the uterus, bladder, or prostatic fossa, surgical drainage is an option to remove the deposited fluid. (See 'Possible role of surgical drainage' above.)