Seasonal variation in the prevalence of profound hyponatremia (<125 mmol/l) in patients on admission to an acute hospital in Japan

Abstract

Hyponatremia is a common but important electrolyte disorder mostly complicated with other diseases. Recently, we experienced a case of a hypertensive patient in her seventies taking a thiazide diuretic, who presented with disorientation due to severe hyponatremia (serum sodium level, 104 mmol/l) on admission. Taking this opportunity, patients with profound hyponatremia (<125 mmol/l) on admission were investigated. Patients newly admitted to Matsumoto Medical Center (an acute hospital) were surveyed retrospectively for one year from May 1, 2016 to April 30, 2017. Patients with profound hyponatremia on admission were selected, and their clinical characteristics were evaluated. A total of 108 out of 4223 patients (2.6%; 67 men, 41 women) showed profound hyponatremia, and 101 out of 108 patients were 65 years old and over. The prevalence of profound hyponatremia in the warm season of April to October (3.1%, 76 in 2444 patients) was significantly (p = 0.002, χ2 test) higher than that in the cold season of November to March (1.8%, 32 in 1779 patients). The monthly prevalence tended to correlate with the monthly average temperature of the local area (r = 0.517, p = 0.085). There were six patients treated with thiazide diuretics in the warm season, while there were no such patients in the cold season, not reaching a significant difference (p = 0.240). The present study demonstrated an increased risk of hyponatremia in elderly patients during the warm season. Considering also the relevant literature, health professionals should pay enough attention to thiazide- or drug-induced hyponatremia and the impact of water/salt intake for heat exposure.

Introduction

Hyponatremia (serum sodium level <135 mmol/l) is a common but important electrolyte disorder mostly complicated with other diseases, implying an increased mortality risk in the emergency department ¹ and in community subjects. ² Electrolyte disorders have been often linked to diuretic use, and thiazide diuretics are most likely associated with hyponatremia, especially in elderly patients.^3–5 To obtain a common and holistic view, joint European societies (represented by European Renal Best Practice) have developed the Clinical Practice Guideline on the diagnostic approach and treatment of hyponatremia, which classifies hyponatremia in adults as follows: mild, 130–134 mmol/l; moderate, 125–129 mmol/l; and profound, <125 mmol/l. ⁶

Recently, we experienced a case of a hypertensive patient in her seventies taking a thiazide diuretic, who presented with disorientation due to severe hyponatremia (104 mmol/l) on admission. Taking this opportunity, patients with profound hyponatremia on admission were investigated retrospectively for the previous one year, and seasonal variation in its prevalence was found. Here we report such variation in relation to taking diuretics, referring to the relevant literature.

Text

During the year from May 1, 2016 to April 30, 2017, 4223 patients newly admitted to Matsumoto Medical Center, an acute hospital without wards for paediatrics, obstetrics and orthopaedic surgery in that year, were surveyed. Patients with a serum sodium level on admission <125 mmol/l were selected, and their clinical characteristics (gender, age, body mass index (BMI) and existing diuretic treatment on admission) were evaluated by clinical chart review. The monthly average temperature of Matsumoto City in the basin, where the difference in the average temperature between summer and winter is large, was provided by the Japan Meteorological Agency. This retrospective study was performed in accordance with the Declaration of Helsinki ethical guidelines. Statistical analysis was performed using an χ² test with or without Yates’ correction and Pearson’s correlation analysis, where appropriate, with a significance level at p < 0.05.

Out of 4223 patients, 108 patients (2.6%; 67 men, 41 women) showed their serum sodium levels <125 mmol/l. Their median age was 80 years old (range 45–100), and 101 out of 108 patients were 65 years old and over. BMI was <18.5 kg/m² in 30 patients, 18.5–24.9 kg/m² in 62, >25 kg/m² in five, and not recorded in 11. The prevalence of patients with profound hyponatremia (<125 mmol/l) in each month is shown in Figure 1, along with the average temperature of the local area. The prevalence of profound hyponatremia in the warm season of April to October (3.1%, 76 in 2444 patients) was significantly (p = 0.002, χ² test) higher than that in the cold season of November to March (1.8%, 32 in 1779 patients). The monthly prevalence tended to correlate with the monthly average temperature of the local area (r = 0.517, p = 0.085). The prevalence of patients with severe hyponatremia (<115 mmol/l), which is potentially associated with substantial morbidity and increased mortality, ⁷ was also significantly (p = 0.017, χ² test) higher in the warm season (0.41%, 10 in 2444 patients) than in the cold season (0 %, 0 in 1779 patients).

Figure 1.

The monthly prevalence of patients with profound hyponatremia (closed circle, %), along with the monthly average temperature (open square, ºC) of the local area.

Similar seasonal variation in the prevalence of hyponatremia has been reported so far. A report from India demonstrated that incidence of hyponatremia (<120 mmlo/l) was higher during the peak southwest monsoon season, suggesting humidity and temperature as contributing factors. ⁸ A report from Switzerland and Austria demonstrated that the prevalence of hyponatremia (<135 mmlo/l) was significantly increased during heat periods. ⁹ The monthly average temperatures ranged from –2.9 to 20.3ºC, which were lower than those in Matsumoto City (–0.2 to 25.2ºC), but the monthly average humidity ranged 66.2 to 87.8%, which was higher than that in Matsumoto City (56 to 81%). Compared with these two reports, an increase in the prevalence of hyponatremia in summer was not so obvious in our study. A study performed in Italy demonstrated that the prevalence of hyponatremia (<135 mmol/l) or severe hyponatremia (<120 mmol/l) in summer was markedly increased in elderly groups (aged over 65 years).^10,11 The authors described that decline in renal function, salt loss, reduced salt intake and increased water ingestion could all contribute to developing severe hyponatremia in elderly patients during the summer. Our study also demonstrated that 93.5% of patients with profound hyponatremia were 65 years and older, confirming the susceptibility of elderly patients.

The number of patients with profound hyponatremia who had been taking diuretics is given in Table 1, according to the kind of diuretics in the warm or cold season. Overlapped patients, who had been treated with two or more kinds of diuretics, were 12 patients in the warm season and four patients in the cold season. Six and no patients were treated with thiazide diuretics in the warm and cold seasons, respectively, not reaching a significant difference. Influences of diuretics on hyponatremia have been also documented in relation to seasonal variation. Risk of hyponatremia induced by drugs including diuretics as adverse drug reactions was demonstrated to increase during high temperatures. ¹² A study on diuretic therapy-associated electrolyte disorders presenting to the emergency room revealed that thiazide and potassium-sparing diuretics were associated with hyponatremia, while loop diuretics were associated with hypernatremia, by using multivariable logistic regression analysis. ⁵ Such different associations seem to be explained by different diuretic actions on renal tubules in combination with the effect of antidiuretic hormone.^4,5 Our study suggests that use of thiazide diuretics could contribute to provoking profound hyponatremia in the warm season, while a study from China (containing a subtropical climate) found no seasonal variations in thiazide-induced hyponatremia. ¹³ The difference may be attributable to the different climates, taking into account the report on the association of drug-induced hyponatremia and high temperatures. ¹²

Table 1.

Prevalence of patients with profound hyponatremia in the warm or cold season.

	Warm (Apr to Oct)	Cold (Nov to Mar)	p-value
Profound hyponatremia	76 in 2444 (3.1)	32 in 1779 (1.8)	0.002
Patients who had been taking diuretics
Loop diuretic	17 in 76 (22.4)	5 in 32 (15.6)	0.594
Potassium-sparing diuretic	12 in 76 (15.8)	6 in 32 (18.8)	0.925
Thiazide diuretic	6 in 76 (7.9)	0 in 32 (0)	0.240
Vasopressin receptor antagonist	1 in 76 (1.3)	0 in 32 (0)	0.654

The number of patients (%).

Conclusions

The present study demonstrated an increased risk of hyponatremia in elderly patients during the warm season, which seemed to be partially involved in the use of thiazide diuretics. Considering also the relevant literature, health professionals should pay enough attention to thiazide- or drug-induced hyponatremia and the impact of water and salt intake for heat exposure.