Analysis of the Clinical Characteristics of Hyponatremia Induced by Trimethoprim/Sulfamethoxazole.

BACKGROUND: Trimethoprim-sulfamethoxazole (TMP/SMX) causes hyperkalemia, and hyponatremia caused by TMP/SMX is a challenge for clinicians. We described the clinical features of hyponatremia induced by TMP/SMX after collecting cases.
SUMMARY: The median age of the 24 patients (10 males and 14 females) was 67 years (range: 28-90 years). Hyponatremia induced by TMP/SMX manifested as nausea (41.7%) and vomiting (29.2%) or asymptomatic hyponatremia (20.8%). The median duration of hyponatremia was 5 days (range: 3-10 days). The median serum sodium concentration was 118 mmol/L (range: 101-128.1 mmol/L). The serum sodium levels gradually returned to the normal range at 4 days (median; range: 2-14 days) after withdrawing TMP/SMX. KEY MESSAGES: TMP/SMX-induced hyponatremia is a rare and serious adverse reaction. Clinicians should be aware of electrolyte disturbances caused by TMP/SMX and should always consider electrolyte monitoring.

Introduction

Hyponatremia, defined as serum sodium concentrations <135 mmol/L, is a common electrolyte disturbance encountered in clinical practice, occurring in up to 42.6% of hospitalized patients [1]. Hyponatremia can be caused by multiple etiologies with different pathophysiological mechanisms. Drugs are the most frequent causes of electrolyte abnormalities, including hyponatremia. Many medications are related to hyponatremia, such as diuretics, antidepressants, antiepileptics, antibiotics, newer antihypertensive agents, and proton pump inhibitors [2]. Hyponatremia is independently associated with an increased risk of in-hospital mortality [3, 4]. Severe hyponatremia can result in cerebral edema, seizures, coma, and eventually death [5]. However, many patients with hyponatremia are asymptomatic and the condition may be life-threatening if left untreated.

Trimethoprim-sulfamethoxazole (TMP/SMX) is a synthetic sulfonamide antibacterial combination product containing a ratio of 5:1 of SMX to trimethoprim. SMX is a structural analog of para-aminobenzoic acid that competitively binds to the enzyme dihydropteroate synthetase, inhibiting the conversion of para-aminobenzoic acid to dihydrofolic acid. TMP impedes the conversion of dihydrofolic acid to tetrahydrofolic acid by binding to the enzyme dihydrofolate reductase [6]. TMP/SMX blocks two steps in the bacterial biosynthesis of essential nucleic acids and proteins to kill bacteria. With increasing bacterial resistance, the role of TMP/SMX has changed. TMP/SMX continues to be the drug of choice for prophylaxis and the treatment of human immunodeficiency virus-related infections and various less common bacterial infections, including Pneumocystis carinii, Nocardia, toxoplasmosis, and Tropheryma whippelii infections [7]. TMP/SMX is available in oral and intravenous preparations and is generally well tolerated, with adverse reactions occurring in 6–8% of individuals [7]. Gastrointestinal and cutaneous symptoms are the most commonly encountered adverse effects. TMP/SMX causes hyperkalemia, particularly at a higher dosage [8]. The first case of hyponatremia caused by TMP/SMX was reported in 1984 [9]. Hyponatremia induced by TMP/SMX has been reported occasionally but is often less highlighted by clinicians; its prevalence is unknown. In this study, we analyzed and discussed the clinical characteristics of TMP/SMX-induced hyponatremia, providing a reference for the safe and reasonable clinical application of TMP/SMX.

Methods

Search Strategy

We searched PubMed/Medline, Web of Knowledge, the Evidence-based Medicine database, Elsevier, Springer Link, Embase, Cochrane Library, China National Knowledge Infrastructure (CNKI), Wanfang Data, and China Science and the Technology Journal Database in Chinese and English from 1984 to 2020. We identified the literature regarding hyponatremia induced by TMP/SMX using the search terms “hyponatremia,” “trimethoprim-sulfamethoxazole,” “trimethoprim,” “TMP/SMX,” “TMP,” “sulfamethoxazole,” “syndrome of inappropriate antidiuretic hormone secretion,” “SIADH,” “electrolyte disturbance,” and “renal salt wasting.”

Inclusion and Exclusion Criteria

The inclusion criteria were as follows: We collected case reports of TMP/SMX-induced hyponatremia for inclusion as preliminary studies. (1) The diagnosis of hyponatremia was clear and related to TMP/SMX. (2) These studies had complete clinical data, including basic patient information, usage and the dosage of TMP/SMX, clinical symptoms, laboratory examination, treatment and outcome. The exclusion criteria were as follows: duplicate literature, reviews, mechanism studies, animal studies, and studies with only an abstract no full text.

Survey Items

Two researchers independently conducted a preliminary screening of the literature according to the inclusion and exclusion criteria, and then the group discussed the literature to be included in the analysis. We used a self-designed data extraction table to extract the following patient information: nationality, sex, age, primary disease, combined disease, TMP/SMX application, combined medication, clinical symptoms, laboratory examination, and disposal and prognosis.

Statistical Analysis

Statistical analyses were performed using SPSS 22.0 (SPSS Inc., Chicago, IL, USA). The results were expressed as medians with ranges and percentages (%).

Results

Patient Characteristics

We identified and screened 1,276 studies using the search strategy described in the Methods section. After careful review by 2 independent authors, only 21 studies were included according to the inclusion and exclusion criteria (Table 1) [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29]. The basic information of 24 patients (10 male and 14 female) is summarized in Table 2. The patients were mainly Americans (37.5%) and Asians (41.7%). The median age of these patients was 67 years (range: 28–90 years). TMP/SMX was mainly used to treat Pneumocystis and Nocardia pneumonia and urinary tract infections caused by Escherichia coli. The daily dose of TMP/SMX was mainly <10 mg/kg/day in 11 patients (52.4%) and 15–20 mg/kg/day in 8 patients (38.1%). The median time of hyponatremia was 5 days (range: 3–10 days). Fifteen patients (62.5%) used medications that caused electrolyte disturbances at the same time, such as steroid therapy in 7 patients (46.7%), antidepressant drugs, angiotensin-converting enzyme inhibitors, or angiotensin receptor blockers and diuretics in 3 patients (20.0%). Twenty patients (83.3%) had underlying diseases, such as hematologic neoplasms, cancer, autoimmune disease, acquired immunodeficiency syndrome, endocrine system diseases, central nervous system diseases, cardiovascular diseases, and respiratory diseases. Two patients developed hyponatremia within 3 days after the readministration of TMP/SMX [14, 16].

Table 1

Basic data of the 24 included patients

Reference	Country	Age/sex	Indication	Dosage	Time, days	Clinical symptoms	Serum sodium (before/after)	Serum osmolality, mOsm/kg	Urine sodium	Urine osmolality, mOsm/kg	Treatment
[9]	UK	75/f	UTI	200 mg bid*	4	Nausea, anorexia	136/107	225	31	523	Discounted, fluid restriction
[10]	Sweden	83/f	Pneumonia (A. xylosoxidans)	160/800 mg iv q8 h	5	Nausea	134/110	242	117	478	Discounted, salt tabs, tolvaptan
[11]	Japan	57/f	Prophylaxis	160/800 mg po qd	3	NR	143/125	264	92	510	Discounted
[12]	USA	54/f	Sinus infection	160 mg po bid	3	Nausea, vomiting, fatigue	NR/101	225	128	128	Discounted, fluid restriction, NS
[13]	Iran	55/f	PP	15 mg/kg iv	10	Nausea, vomiting, confused, lethargic	146/110	240	50	NR	Discounted, HS, restore after 2 d
[14]	Germany	66/m	Prophylaxis	960 mg po bid	NR/3	Tremo	NR/104	226	NR	238	Discounted
[15]	India	49/m	UTI (E. coli)	20 mg/kg/day po*	7	Fatigue, nausea, vomiting	NR/115	248	142	161	Discounted, high salt diet, restore after 4 d
[16]	USA	82/f	UTI (GBS)	160/800 mg po bid	7/3	Depression, insomnia, nausea	132/121	NR	NR	NR	Discounted, NS, restore after 8 d
[17]	Australia	69/f	PJP	320/1600 mg iv q8 h	5	Confused	NR/118	248	135	398	Discounted, HS, restore after 5 d
[17]	Australia	90/f	UTI (E. coli)	300 mg po qd*	4	Nausea, vomiting, delirium	NR/123	255	107	377	Discounted, restore after 4 d
[17]	Australia	75/m	UTI	300 mg po qd*	5	ACS	NR/123	264	118	411	Discounted, NS
[18]	USA	28/m	PJP	15 mg/kg/d iv*	5	Tachycardia	135/118	245	136	682	Discounted, salt tabs, tolvaptan, NS, restore after 2 d
[19]	USA	86/m	UTI	80/400 mg po bid	4	AMS, weakness, fatigue	NR/127	275	86	659	Discounted, fluid restriction, NS, restore after 8 d
[20]	Israel	78/f	UTI	NR	7	Weakness, abdominal pain, nausea, vomiting	NR/115	250	114	457	Discounted
[21]	UK	53/f	PCP	20 mg/kg po*	12	Weakness	137/118	NR	41	287	Discounted, restore after 7d
[22]	Japan	64/m	PCP	12 g/d po	4	ACS	139/109	NR	NR	NR	Continued, HS, fluid restriction
[22]	Japan	59/m	PCP	12 g/d iv	6	NR	141/126	NR	NR	NR	Discounted, HS, fluid restriction
[23]	USA	42/m	PCP	20 mg/kg/day iv*	4	NR	137/121	NR	142	706	Discounted
[24]	USA	64/f	Pneumonia	16 mg/kg/day iv*	5	NR	NR/128	NR	NR	NR	Discounted, restore after 3d
[25]	USA	68/m	Brain abscesses (Nocardia)	15 mg/kg iv q8 h	3	Abdominal pain, nausea, vomiting	136/124	261	53	235	Discounted, salt tablets, restore after 2 d
[26]	USA	83/f	Pneumonia (A. xylosoxidans)	160/800 mg iv q8 h	3	NR	NR/110	242	114	478	Continued, tolvaptan, fluid restriction, salt tablets
[27]	Turkey	65/f	PCP	15 mg/kg iv q6 h*	10	NR	136/120	254	170	529	Discounted, fluid restriction, HS, restore after 4 d
[28]	USA	72/m	Pneumonia (N. cyriacigeorgica)	480/2400 mg po bid, 10 mg/kg/day*	5	Confused	133/112	245	109	422	Discounted, fluid restriction, salt tablets, restore after 14 d
[29]	China	75/f	UTI (E. coli)	80/400 mg po bid	6	Nausea, vomiting	140.6/128.1	NR	NR	NR	Continued, HS, sodium bicarbonate

ACS, altered conscious state; AMS, altered mental status; A. xylosoxidans, Achromobacter xylosoxidans; E. coli, Escherichia coli; GBS, Group B Streptococcus; HS, hypertonic saline; IV, intravenous; NS, normal saline; N. cyriacigeorgica, Nocardia cyriacigeorgica; NR, not reported; PCP, Pneumocystis carinii pneumonia; PJP, Pneumocystis jirovecii pneumonia; PP, Pneumocystis pneumonia; UTI, urinary tract infection.

Represents the trimethoprim dose.

Table 2

Summary of the basic information of the 24 included patients

Parameter	Value, n (%)
Sex (24)b
Female	14 (58.3)
Male	10 (41.7)
Age (24)b
Year	67 (28.90)a
Region (24)b
USA	9 (37.5)
Japan	3 (12.5)
Australia	3 (12.5)
UK	2 (8.3)
Iran	1 (4.2)
India	1 (4.2)
Sweden	1 (4.2)
Israel	1 (4.2)
Turkey	1 (4.2)
Germany	1 (4.2)
China	1 (4.2)
Indication (24)b
Pneumonia	12 (50.0)
Urinary tract infection	8 (33.3)
Prophylaxis	2 (8.3)
Sinus infection	1 (4.2)
Brain abscesses	1 (4.2)
Bacterium (17)b
Pneumocystis	8 (47.1)
Pneumocystis carinii	5 (62.5)
Pneumocystis jirovecii	2 (25.0)
Nocardia	3 (17.7)
Nocardia cyriacigeorgica	1 (33.3)
Nocardia farcinica	1 (33.3)
Escherichia coli	3 (17.7)
Achromobacter xylosoxidans	2 (11.8)
Group B Streptococcus	1 (5.9)
Daily dose of TMP/SMX (21)b
15–20 mg/kg/day	8 (38.10)
10–15 mg/kg/day	2 (9.5)
<10 mg/kg/day	11 (52.4)
Route of administration (24)b
Oral	13 (54.2)
Intravenous	11 (45.8)
Onset time of hyponatremia (24)b
Days	5 (3.10)b
Underlying illness (20)b
Solid cancer	3 (15.0)
Hematologic neoplasms	2 (10.0)
Kidney transplant	2 (10.0)
Autoimmune disease	5 (10.0)
AIDS	3 (15.0)
Endocrine system diseases	4 (20.0)
Central nervous system diseases	4 (20.0)
Cardiovascular diseases	9 (45.0)
Respiratory diseases	4 (20.0)
Concomitant medication (15)b
Steroid therapy	7 (46.7)
Antidepressant drugs	3 (20.0)
ACEI/ARB	3 (20.0)
Diuretics	3 (20.0)

TMP/SMX, trimethoprim-sulfamethoxazole; AIDS, acquired immunodeficiency syndrome; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker.

Median (minimum-maximum).

Represents the number of patients among a total of 24 on whom information regarding this particular parameter was provided.

Clinical Manifestations

The clinical manifestations of hyponatremia induced by TMP/SMX are summarized in Table 3. Nausea in 10 patients (41.7%) and vomiting in 7 patients (29.2%) were the most common clinical symptoms of TMP/SMX-induced hyponatremia, followed by weakness and fatigue in 5 patients (20.8%), confusion in 3 patients (12.5%), and an altered conscious state in 3 patients (12.5%). Six patients (25.0%) presented with asymptomatic hyponatremia.

Table 3

Clinical information of the 24 included patients

Parameter	Clinical features	Value, n (%)	Normal range
Presenting symptoms (24)b	Nausea	10 (41.7)
	Vomiting	7 (29.2)
	Fatigue and weakness	5 (20.8)
	Anorexia	2 (8.3)
	Tachycardia	2 (8.3)
	Confused	3 (12.5)
	Altered conscious state	3 (12.5)
	Delirium	1 (4.2)
	Lethargic	1 (4.2)
	Abdominal pain	2 (8.3)
	Tremor	1 (4.2)
	Depression	1 (4.2)
	Asymptomatic	6 (25.0)
Sodium	Before medication (14)b	136.5 (132, 146)a	135–145 mmol/L
	After medication (24)b	118 (101, 128.1)a
	130–135 mmol/L	0 (0)
	125–130 mmol/L	4 (16.7)
	≤125 mmol/L	20 (83.3)
Potassium	Before medication (8)b	4.2 (2.7, 4.96)a	3.5–5.5 mmol/L
	After medication (18)b	5.3 (3.1, 6.6)a
Creatinine	Before medication (12)b	0.85 (0.11, 1.6)a	0.7–1.2 mg/dL
	After medication (17)b	1.1 (0.47, 4)a
Serum osmolarity (18)b		248 (225, 275)a	285–295 mOsm/kg
Bicarbonate (9)b		17.3 (8, 24)a	22–29 mmol/L
Urine sodium (17)b		114 (31, 170)a	<20 mmol/L
Urine osmolality (18)b		439.5 (128, 706)a	50–1,400 mOsm/kg
FENa (10)b		1.61 (0.36, 6.5)a	<1%

FENa, fractional excretion of sodium.

Serum sodium, potassium, and creatinine in parentheses represent the median value (range: minimum-maximum) before and after administration.

Represents the number of patients among a total of 24 on whom information regarding this particular parameter was provided.

Laboratory Tests

The laboratory test results are described in Table 2. The median serum sodium concentration of hyponatremia induced by TMP/SMX was 118 mmol/L (range: 101–128.1 mmol/L), and the median serum potassium concentration was 5.3 mmol/L (range: 3.1–6.6 mmol/L). Twenty patients (83.3%) developed severe hyponatremia (≤125 mmol/L). The median plasma osmolality was 248 mOsm/kg (range: 225–275 mOsm/kg). The median level of serum bicarbonate was 17.3 mmol/L (range: 8–24 mmol/L). Urine studies showed that the median urine osmolality was 439.5 mOsm/kg (range: 128–706 mOsm/kg), the median urine sodium was 114 mmol/L (range: 31–170 mmol/L), and the median urinary fractional excretion of sodium was 1.61%. Acute kidney injury occurred in 3 patients (12.5%) [24, 25, 28]. Renin and aldosterone were elevated in 4 patients [11, 13, 18, 27].

Treatment and Prognosis

Twenty patients (83.3%) immediately stopped TMP/SMX when hyponatremia occurred (Table 4). Normal saline was administered to 7 patients (29.2%), hypertonic saline to 8 patients (25.0%), fluid restriction to 8 patients (33.3%), salt tablets to 4 patients (16.7%), and a high-salt diet to 1 patient (4.2%). Three patients (12.5%) were given high-dose tolvaptan (≥30 mg/day) to treat hyponatremia [19, 23, 24]. The serum sodium levels of these patients gradually returned to the normal range at a median time of 4 days (range: 2–14 days).

Table 4

Treatment of the 24 included patients

Parameter	Value, n (%)
Therapy (24)a
Discontinued	20 (83.3)
Continued	4 (16.7)
Normal saline	7 (29.2)
Hypertonic saline	6 (25.0)
Fluid restriction	8 (33.3)
Salt tablets	4 (16.7)
Tolvaptan	3 (12.5)
High salt diet	1 (4.2)
Recovery time of serum sodium to normal ranges (12)b, days	4 (2.14)a

TMP/SMX, trimethoprim-sulfamethoxazole.

Median (minimum–maximum).

Represents the number of patients among a total of 24 on whom information regarding this particular parameter was provided.

Discussion

Hyponatremia occurs frequently in hospitals and usually brings challenges in diagnosis or treatment. TMP/SMX-induced hyponatremia is often misdiagnosed as a syndrome of inadequate antidiuretic hormone secretion (SIADH). The exact mechanism of TMP/SMX-induced hyponatremia is unclear. Current research suggests that renal salt wasting may be the main underlying mechanism of hyponatremia caused by TMP/SMX [18, 30]. TMP is similar in structure to amiloride and acts as a potassium-sparing diuretic by blocking the epithelial sodium channel in the collection tube, resulting in natriuresis, hyperkalemia, and metabolic acidosis [31]. A retrospective study found that 72.3% of hospitalized patients who had used high-dose TMP/SMX (8 mg/kg/day of TMP for ≥3 days) had hyponatremia, with the severity of hyponatremia depending on longer duration and higher cumulative doses of TMP/SMX [32]. Mori et al. [33] retrospectively found that patients treated with TMP/SMX will also have electrolyte disturbances even with a low dose (TMP: <80 mg) or standard dose (TMP: 80–120 mg), indicating a dose-dependent effect. Generally, electrolyte disturbances were observed more frequently with high-dose TMP/SMX, and low doses can also cause both hyperkalemia and hyponatremia.

Nervous system symptoms are the main clinical manifestations of hyponatremia, ranging from nausea, vomiting, headache and general malaise to changes in mental status, seizures, and even coma [12]. Our study showed that TMP/SMX-induced hyponatremia mainly manifests as nausea, vomiting, weakness, and fatigue and can also manifest as asymptomatic hyponatremia. TMP/SMX mainly causes severe hyponatremia (serum sodium ≤125 mmol/L) with or without elevated potassium. TMP may cause a distal acidification defect, with hyperkalemia contributing to the development of acidosis through effects on the renal handling of ammonia and aldosterone secretion. This leads to a decreased serum bicarbonate level, which is consistent with our clinical findings [34]. Renin/aldosterone tends to be suppressed in SIADH and elevated in renal salt wasting [18]. Urinary sodium excretion of hyponatremia is related to the waste of kidney salt> 20 mmol/L [35].

Previous studies have found that African-American patients, elderly individuals, patients with renal disease, and patients with diuretics have an increased risk of developing hyponatremia [19, 36]. Stephens et al. [37] found that approximately one-third of children aged <18 years receiving intravenous SMP/TMX developed hyponatremia, and concomitant furosemide administration was one of the most common risk factors. SMP/TMX combined with systemic corticosteroids with mineralocorticoid effects may compensate for TMP-related hyponatremia [11]. A case-control study showed that the simultaneous use of TMP/SMX and inhibitors of the renin-angiotensin system increases the risk of hyperkalemia. These results suggest that ACEIs/angiotensin receptor blockers can amplify the adverse effects of TMP on the renal distal tubule [38]. Caution should be taken when TMP/SMX is combined with hyponatremic-inducing drugs, such as serotonin receptor uptake inhibitors [28]. Some patients with hyponatremia have a complex and multifactorial etiology. A thorough investigation should be conducted to ensure the correct diagnosis, particularly when diseases that may cause hyponatremia, such as malignancy, adrenal insufficiency, and hypothyroidism, are present [39, 40]. Human immunodeficiency virus infection may be associated with SIADH, with previous studies reporting SIADH in up to 53% of hospitalized patients with acquired immune deficiency syndrome [41].

TMP/SMX-induced hyponatremia can be resolved by discontinuing the drug, intravenous saline hydration, a high-salt diet or 3% saline. The effectiveness and safety of vasopressin receptor antagonists in treating symptomatic or acute hyponatremia are not yet well established. However, a few studies have proven that a massive dose of tolvaptan (30 mg) can effectively treat hyponatremia caused by TMP/SMX in patients who are refractory to salt tablets and fluid restriction [10]. Our study found that the serum sodium concentration returned to the normal range approximately 4 days after stopping TMP/SMX. Additionally, Huntsberry et al. [16] and Herzog et al. [14] found that hyponatremia recurred after re-exposure to TMP/SMX with a shorter onset time [14, 16].

Conclusion

Hyponatremia is a serious and rare adverse reaction to TMP/SMX. Various doses of TMP/SMX may lead to hyponatremia. Close monitoring of serum sodium should be performed in patients who are receiving TMP/SMX therapy, particularly in patients with high-risk factors. Hyponatremia should be considered in patients on TMP/SMX who present with digestive or neurological symptoms.

Conflict of Interest Statement

The authors declare no conflicts of interest.

Funding Sources

This study was funded by the Natural Science Foundation of Hunan Province (No. 2021JJ80083).

Author Contributions

Chunjiang Wang and Haibo Lei designed the study, Xiang Liu provided clinical support, Jiang Zeng and Zhiqiang Fan collected the data, Zuojun Li and Yang He analyzed the data, Haibo Lei and Chunjiang Wang wrote the manuscript (original draft preparation), Xiang Liu and Jiang Zeng wrote the manuscript (review and editing), and Zhiqiang Fan and Zuojun Li provided conceptual advice. All the coauthors have agreed to the submission of the final manuscript.