Recurrent Severe Hyponatremia Following Polyethylene Glycol Electrolyte Lavage Solution with Ascorbic Acid in a Patient with Undiagnosed Syndrome of Inappropriate Antidiuretic Hormone Secretion

Introduction

Colonoscopy is the criterion standard for detecting colorectal cancer [1], and effective mucosal evaluations require adequate bowel preparation. Among the available bowel cleansing agents, polyethylene glycol electrolyte lavage solution (PEG-ELS) is widely used because of its nonreactive, water-soluble nature and minimal intestinal absorption [2,3]. Although PEG-ELS is generally safe, it can cause adverse effects, such as renal failure, hyponatremia, upper gastrointestinal bleeding, aspiration attributable to vomiting, and death [4–6]. Hyponatremia is a well-documented adverse event linked to bowel preparations that particularly affects vulnerable populations, such as elderly individuals. Hyponatremia cases are often attributed to excessive water intake, impaired renal water excretion, or inappropriate antidiuretic hormone secretion (SIADH) [7] potentially triggered by bowel preparation agents.

We present the case of a patient with serious hyponatremia induced by PEG-ELS with ascorbic acid (PEG-ELS/Asc) who received a diagnosis of SIADH, based on exclusion criteria. Prostate cancer was subsequently investigated as a potential underlying cause, following the SIADH diagnosis. Although hyponatremia following bowel preparation has been reported, this case highlights the reproducibility of symptoms with PEG-ELS/Asc and underscores the clinical relevance of careful agent selection. Such reproducible cases are rare and underscore the importance of risk management for elderly patients.

Case Report

A 79-year-old man presented to our hospital for endoscopic treatment of a colorectal tumor. His medical history included hypertension, which was managed with candesartan and amlodipine, and diabetes mellitus, which was managed with glibenclamide, sitagliptin, and metformin. On the day of the scheduled colonoscopy, the patient completed bowel preparation consisting of PEG-ELS/Asc during the morning. The patient ingested 2000 mL of PEG-ELS/Asc and was instructed to drink 1000 mL of free water, resulting in a total of approximately 3000 mL of ingested fluid. However, stupor and generalized tonic-clonic seizures developed 5 h after PEG-ELS/Asc intake.

The blood pressure and heart rate of the patient were 129/85 mmHg and 82 beats/min, respectively. A physical examination revealed no evidence of jugular venous distention, leg edema, decreased skin turgor, or dry mucous membranes, thus indicating a normal volume status. A neurological examination revealed no obvious focal neurological deficits.

Laboratory test results (Table 1) indicated a serum sodium (Na) level of 119 mEq/L, reduced total serum osmolality of 255 mOsm/L (<275 mOsm/L), and elevated urine osmolality of 418 mOsm/L (>100 mOsm/L). The serum creatinine level was 0.64 mg/dL, and the blood urea nitrogen level was 14.0 mg/dL. Additional testing confirmed an antidiuretic hormone secretion (ADH) level of 2.6 pg/mL (reference range, 0.0–6.7 pg/mL), adrenocorticotropic hormone level of 20.2 pg/mL (reference range, 7.2–63.3 pg/mL), cortisol level of 17.1 mg/dL (reference range, 7.07–19.60 mg/mL), thyroid-stimulating hormone level of 2.83 μIU/mL (reference range, 0.5–5 μIU/mL), and free thyroxine level of 1.38 ng/dL (reference range, 0.9–1.7 ng/dL). Computed tomography (CT) and magnetic resonance imaging (MRI) of the brain performed the following day revealed chronic ischemic changes and age-related brain atrophy; however, neither active changes nor acute changes were observed (Figure 1).

Acute hyponatremia was diagnosed, and 3% hypertonic NaCl solution was initially administered intravenously. This treatment resulted in an improved neurological status, including the ability to follow simple commands and move all limbs equally. After 48 h of treatment with continuous intravenous infusion of 3% hypertonic NaCl solution, the patient began to respond to vocal stimulation and his condition returned to normal; no residual neurological abnormalities were observed.

The laboratory data showed that urine osmolality shifted from 418 mOsm/kg at the onset of serious hyponatremia (serum Na level, 119 mEq/L) to 132 mOsm/kg after 3% NaCl (serum Na level, 122 mEq/L) administration. The urine creatinine-to-serum creatinine ratio was 20.16 (urine creatinine, 12.9; serum creatinine, 0.64) [8]. These data suggest that decreased free water excretion is associated with excess ADH; however, prompt administration of hypertonic NaCl stimulated the excretion of hypotonic urine and immediately resolved the neurological and gastrointestinal symptoms. Electrolyte levels remained normal until discharge and during subsequent outpatient follow-up. A blood examination, gastrointestinal endoscopy, and CT of the chest and abdomen were performed to investigate the underlying cause of SIADH and revealed elevated prostate-specific antigen levels; additionally, abdominal CT and MRI suggested prostate cancer with para-aortic lymph node metastasis (Figure 2).

A subsequent prostate biopsy confirmed the presence of prostate cancer, which is a potential cause of SIADH that was diagnosed through exclusion criteria. Hormonal therapy for prostate cancer was initiated after the diagnosis was confirmed by a urologist. The colonic tumor was diagnosed as intramucosal carcinoma, without invasive or metastatic features. Therefore, endoscopic mucosal resection was indicated 2 months later. The diagnosis of SIADH was confirmed before colonoscopy. The patient was treated as an outpatient and prescribed 3 g of oral NaCl per day for 10 weeks. Fluid restriction was not implemented prior to the second colonoscopy. For this patient, we decided to supplement NaCl and not restrict water. We believe that elderly patients are not able to cope with fluctuations in the fluid volume in the body that could cause adverse events associated with water withdrawal. Additionally, because a colonoscopy was scheduled, the risk of further rapid fluctuations in fluid delivery associated with the use of intestinal lavage fluid was possible. The patient was readmitted to the hospital the day before the scheduled colonoscopy after it was confirmed that he was asymptomatic and did not have a decreased serum Na level. The patient completed bowel preparation using PEG-ELS/Asc. However, after the colonoscopy was performed, neurological symptoms and muscle spasms caused by hyponatremia developed again and were similar to those during the previous episode. Therefore, the aforementioned treatment measures were implemented again, resulting in improved symptoms (Table 2). Hyponatremia and SIADH recurrences were associated with bowel preparation, thus demonstrating reproducibility.

We asked the patient’s former physician about the bowel preparation method used before the previous colonoscopy. We were informed that the patient used PEG-ELS without Asc, and that obvious symptoms suggestive of hyponatremia did not occur. Based on this information, we concluded that the patient had undiagnosed SIADH, with prostate cancer detected through systematic cancer screening, which occurred with PEG-ELS/Asc use but not with PEG-ELS use for bowel preparation (Table 3). The patient’s symptoms were reproduced during 2 separate events; therefore, PEG-ELS/Asc was identified as the trigger.

Discussion

We encountered a case of significant symptomatic hyponatremia after bowel preparation using PEG-ELS/Asc. Proper bowel preparation is essential before colonoscopy. When choosing a bowel preparation strategy, safety is an important consideration. Although PEG-ELS/Asc is generally considered safe, this case necessitated further evaluation.

Risk factors for hyponatremia following bowel preparation include age older than 65 years, female sex, renal insufficiency, electrolyte abnormalities, heart failure, excessive fluid consumption, gastrectomy, angiotensin-converting enzyme inhibitor use, angiotensin II receptor blocker use, thiazide diuretic use, and antidepressant use [9–13]. The prevalence of hyponatremia induced by bowel preparation ranges from 1% to 7% [10,11,14]. The occurrence of hyponatremia following bowel preparation can manifest through various physiological pathways. A decrease in fluid volume during bowel preparation can lead to increased secretion of ADH from the posterior pituitary gland [10,13,15]. Additionally, protein and Na deficiencies caused by fasting during the bowel preparation period can play a role in hyponatremia development [16]. Moreover, patients who consume substantial amounts of water to compensate for fluid loss through the colon during the bowel preparation period can be at higher risk for hyponatremia [9,10,13,16]. Colonoscopy can trigger ADH release [17–19]. Nevertheless, most colonoscopy strategies do not lead to hyponatremia, even in elderly patients.

The diagnostic criteria for SIADH, which was initially described by Bartter and Schwartz in 1967, are widely known [20]. For patients with SIADH, hyponatremia occurs as a result of ADH-induced retention of ingested or infused water, with intact Na handling. Abnormalities of volume-regulating mechanisms, such as the renin-angiotensin-aldosterone system and atrial natriuretic peptide, were not observed [21]. SIADH has various causes, including ectopic production of arginine vasopressin by neoplasms, acute infections, pulmonary diseases, and central nervous system disorders. Additionally, SIADH has been associated with various malignancies [22]. Arginine vasopressin is often synthesized in tumor tissues [23], thus causing ectopic hormone secretion. Although SIADH in patients with cancer has multiple causes, the cause of hyponatremia is often multifactorial. Therefore, a systematic evaluation is necessary to define the etiology of hyponatremia before colonoscopy. Additionally, copeptin, the C-terminal segment of the arginine vasopressin prohormone, is an arginine vasopressin surrogate with high ex vivo stability that is easy to measure; therefore, it can be useful for distinguishing hypotonic hyponatremia. However, copeptin measurement is not performed in general laboratories in Japan [24].

In the present case, undiagnosed SIADH was revealed during colonoscopy, and a subsequent examination revealed prostate cancer. SIADH occurred with PEG-ELS/Asc use at our hospital; however, a previous colonoscopy at another facility that included bowel preparation using PEG-ELS without Asc did not result in hyponatremia. The type of bowel preparation used and the patient’s risk factors can influence the development of hyponatremia. PEG-ELS/Asc and other PEG-based solutions have been implicated in several cases of hyponatremia.

In clinical practice, PEG-ELS is considered an isotonic and nonabsorbable osmotic laxative that rarely causes significant electrolyte imbalances. However, the addition of Asc to PEG-ELS – as in PEG-ELS/Asc formulations – can substantially alter osmotic dynamics and fluid-electrolyte homeostasis, leading to an increased risk of hyponatremia.

The pathophysiological mechanisms underlying this phenomenon include the osmotic load associated with unabsorbed Asc, dilutional hyponatremia caused by free water intake, and ADH stimulation by metabolic acidosis. High doses of orally administered Asc exceed the intestinal absorption capacity, particularly in the jejunum. When the absorption of high-concentration Asc in the intestine reaches saturation (the daily intake is more than 1 g), the surplus remains in the intestinal cavity up to the colon and can act synergistically with PEG as an osmotic laxative [25]. This results in increased intraluminal osmotic pressure and the drawing of water from the vascular compartment into the intestinal lumen. Although this osmotic diarrhea contributes to effective bowel cleansing, it also leads to significant free water loss and relative Na depletion. Because of dilutional hyponatremia caused by free water intake, patients are commonly instructed to consume large volumes of water (eg, 500 mL per 1 L of PEG-Asc solution) to support bowel cleansing. However, when this fluid is hypotonic (eg, plain water), excessive intake without proportional Na replacement can result in dilutional hyponatremia. For elderly patients and those with impaired renal water excretion, this risk is particularly increased [12]. In the liver and kidneys, Asc is metabolized to dehydroascorbic acid and oxalate, potentially causing mild metabolic acidosis in susceptible individuals. In fact, high doses of Asc administration can produce acidic metabolites in the body (especially oxalic acid), thus increasing the risk of renal dysfunction and metabolic acidosis [26–28]. This acid-base disturbance can stimulate ADH secretion, thus leading to water retention through enhanced renal water reabsorption [29]. Consequently, the serum Na concentration is further reduced.

Therefore, unlike PEG-ELS/Asc, standard PEG-ELS solutions maintain isotonicity and contain balanced electrolytes, thus minimizing net osmotic shifts and Na dilution. Additionally, they do not introduce a large acidic osmotic solute, such as Asc; therefore, the risk of hyponatremia is significantly lower.

The high osmolality of PEG-ELS/Asc is effective but can cause rapid changes in the water content of the body. Moreover, before colonoscopy, patients may experience stress and anxiety, which can result in nausea and vomiting. Nausea and vomiting can trigger nonosmotic ADH release, leading to inappropriate water reabsorption and further dilution of serum Na. The hyponatremia and neurological symptoms experienced by our patient were reproduced during 2 separate events; therefore, PEG-ELS/Asc was identified as a trigger. SIADH was diagnosed based on data obtained during the acute phase. An assessment was not performed during the stable phase, when the serum Na level was normal. Therefore, we concluded that SIADH caused by prostate cancer likely did not have a chronic effect; however, we also concluded that it likely affected acute-onset hyponatremia during bowel preparation. Various complex factors may have contributed to the development of SIADH and hyponatremia in this case. SIADH occurred with PEG-ELS/Asc use by our patient; however, notably, it did not occur with PEG-ELS use. Additionally, our patient was using angiotensin receptor blockers. Therefore, the bowel preparation approach before colonoscopy, especially that used by older patients who may have many risk factors, should be carefully considered. Although water restriction is often the first choice for SIADH treatment, we decided against aggressive water restriction because of the patient’s advanced age and the risk of cardiovascular damage. In retrospect, drinking intestinal lavage solution as well as restricting water were considered risky because they can cause further fluid withdrawal in the body. We also used the same bowel lavage solution because we believed that SIADH could be prevented by salt loading; however, we found this case enlightening and educational because our patient developed the same symptoms twice. We believe that this case report provides important information regarding the selection of enteric lavage solutions.

Therefore, because of the increasing use of PEG-ELS/Asc solutions for clinical bowel preparation, their safety for elderly patients remains a significant concern. Although PEG-ELS/Asc is generally effective for and well-tolerated by the general population, recent studies have highlighted important safety issues for older individuals. Schneider et al demonstrated that excessive intake of clear liquids in addition to PEG-ELS/Asc intake significantly increase the risk of hyponatremia, particularly for patients with an impaired electrolyte regulatory capacity, such as elderly individuals [30]. Moreover, Kang et al showed that, compared with PEG-ELS/Asc, oral sulfate tablets provided superior cleansing quality, higher adenoma detection rates, and better tolerability for patients 70 years or older [31].

These findings suggest that PEG-ELS/Asc should not be recommended as the first-line bowel preparation option for older patients, particularly those with comorbidities, such as chronic kidney disease, heart failure, or polypharmacy involving diuretics. For such individuals, oral sulfate tablets or other low-volume, nonhypotonic agents can be safer alternatives. If PEG-ELS/Asc is used for elderly individuals, then its administration should be carefully controlled; for example, fluid intake must be limited (≤2 L), and serum electrolytes should be monitored, especially in inpatient settings. Risk assessment tools and pre-procedure screening can help identify patients at high risk for electrolyte disturbances. This tailored approach can help maximize procedural safety without compromising bowel preparation efficacy.

Conclusions

Seizures occurred because of severe hyponatremia in an elderly patient with SIADH triggered by the use of PEG-ELS/Asc for bowel preparation. In this case, PEG-ELS/Asc caused hyponatremia, consisting of the same symptoms on 2 occasions. This case is relatively rare; therefore, we focused on Asc to determine the detailed mechanisms underlying hyponatremia development. An appropriate risk assessment is essential before the use of bowel preparation agents. This rare case with replicated symptoms on 2 separate occasions emphasizes the importance of individualized risk management strategies for elderly patients. Physicians should recognize the potential symptoms and adverse effects caused by PEG-ELS/Asc for bowel preparation.