Association between normal saline infusion volume in the emergency department and acute kidney injury in heat stroke patients: a multicenter retrospective study

Figures & data

Figure 1. Flowchart.

Table 1. Characteristics of the patients.

Variables	All (n = 138)	Acute kidney injury		p
		No (n = 105)	Yes (n = 33)
Baseline characteristics
Age (years)	65.26 ± 14.84	66.46 ± 14.57	61.45 ± 15.26	.068
Female, n (%)	55 (39.86)	42 (40.00)	13 (39.39)	.951
Comorbidities, n (%)
Diabetes	9 (6.52)	5 (4.76)	4 (12.12)	.135
Chronic cardiovascular disease	40 (28.99)	31 (29.52)	9 (27.27)	.977
Mental disorder	14 (10.14)	10 (9.52)	4 (12.12)	.666
Dementia	5 (3.62)	4 (3.81)	1 (3.03)	.834
Institution, n (%)				.030
1	47 (34.0)	40 (38.10)	7 (21.21)
2	19 (13.77)	13 (12.38)	6 (18.18)
3	22 (15.94)	17 (16.19)	5 (15.15)
4	15 (10.87)	14 (13.33)	1 (3.03)
5	22 (15.94)	15 (14.29)	7 (21.21)
6	13 (9.42)	6 (5.71)	7 (21.21)
Type of heat stroke, n (%)				.555
Classic	52 (37.68)	41 (39.05)	11 (33.33)
Exertional	86 (62.32)	64 (60.95)	22 (66.67)
Chief complaints, n (%)
Coma	113 (81.88)	82 (78.10)	31 (93.94)	.039
Epileptic seizure	19 (13.77)	12 (11.43)	7 (21.21)	.155
Onset time (h)	1.00 (1.00–3.00)	1.50 (1.00–3.00)	1.00 (1.00–3.00)	.470
Monitoring parameters at admission
Temperature (°C)	40.80 ± 0.81	40.77 ± 0.77	40.89 ± 0.95	.452
Heart rate (beat/min)	128.57 ± 24.08	124.41 ± 23.64	141.82 ± 20.68	<.001
Systolic pressure (mmHg)	126.62 ± 31.54	130.30 ± 31.42	114.94 ± 29.44	.014
Glasgow Coma Scale	8.80 ± 4.19	9.50 ± 4.29	6.58 ± 2.94	<.001
NEWS	12.51 ± 3.59	11.93 ± 3.69	14.36 ± 2.55	<.001
Body temperature after cooling (°C)
Temperature (0.5 h)	39.50 ± 0.99	39.32 ± 0.93	40.01 ± 0.90	<.001
Temperature (2 h)	37.92 ± 1.00	37.73 ± 0.91	38.49 ± 1.05	<.001
Blood test at admission
C-reactive protein (mg/L)	0.62 (0.30–4.40)	0.70 (0.30–5.80)	0.30 (0.30–2.0)	.458
Potassium (mmol/L)	3.64 ± 0.63	3.58 ± 0.60	3.83 ± 0.70	.053
Sodium (mmol/L)	130.46 ± 13.07	130.86 ± 7.67	129.21 ± 23.19	.529
Chloride (mmol/L)	97.54 ± 9.06	97.11 ± 8.93	98.89 ± 9.49	.329
Blood glucose (mmol/L)	11.35 ± 4.87	10.90 ± 4.80	12.70 ± 4.90	.065
Alanine aminotransferase (IU/L)	23 (16.25–35.00)	22.50 (16.25–35)	23 (17.00–32.58)	.620
Creatinine (mmol/L)	98.11 ± 23.35	93.45 ± 23.26	112.92 ± 16.72	<.001
Blood urea nitrogen (mmol/L)	6.86 ± 2.73	6.88 ± 2.82	6.80 ± 2.45	.882
Creatine kinase (U/L)	265.50 (141.80–703.75)	256.00 (133.20–546.55)	395.00(159.50–930.00)	.656
Troponin I (ng/mL)	0.03 (0.01–0.07)	0.02 (0.01–0.06)	0.04 (0.02–0.15)	.061
Prothrombin time (s)	13.25 ± 2.01	13.29 ± 1.90	13.12 ± 2.37	.681
Lactic acid (mmol/L)	3.42 ± 2.00	3.18 ± 2.08	4.15 ± 1.53	.016
White blood cell count (1 × 10^9/L)	10.24 ± 4.71	10.25 ± 4.70	10.23 ± 4.82	.983
Platelet (1 × 10^9/L)	197.01 ± 71.15	194.94 ± 69.07	203.55 ± 78.12	.547
Serum chlorine after normal saline infusion (mmol/L)	107.67 ± 6.96	106.30 ± 6.53	110.83 ± 6.99	.008
Volume of normal saline infusion in ED (L)	3.02 ± 1.45	2.82 ± 1.34	3.66 ± 1.61	.006
Total infusion volume (L)	3.69 ± 1.99	3.48 ± 1.82	4.36 ± 2.19	.043
Use of vasoactive drugs, n (%)	28 (20.29)	13 (12.38)	15 (45.45)	<.001
Length of stay in ED (h)	10.00 (4.50–19.88)	10.00 (4.50–22.00)	8.00 (4.00–16.50	.160
Outcomes
RRT, n (%)	7 (5.07)	2 (1.90)	5 (15.15)	.002
Admitted to ICU, n (%)	52 (37.68)	25 (23.81)	27 (81.82)	<.001
Length of stay in hospital (d)	5.00 (2.00–8.00)	5.00 (0.00–7.00)	7.00 (5.00–12.00)	.005
Length of stay in ICU (d)	0.00 (0.00–4.00)	0.00 (0.00–0.00)	5.00 (3.00–8.00)	<.001
Coagulation impairment, n (%)	20 (14.49)	6 (5.71)	14 (42.42)	<.001
Rhabdomyolysis, n (%)	8 (5.80)	6 (5.71)	2 (6.06)	.941
Mortality, n (%)	10 (7.25)	5 (4.76)	5 (15.15)	.045
Central nervous system damage at discharge, n (%)	16 (11.59)	9 (8.57)	7 (21.21)	.048

NEWS: National Early Warning Score; ED: emergency department; ICU: intensive care unit; RRT: renal replacement therapy.

Table 2. Logistic regression of normal saline infusion volume in emergency department for acute kidney injury in patients with heat stroke.

Volume of normal saline infusion	Acute kidney injury
	Univariate		Adjust model I		Adjust model II
	OR (95% CI)	p	OR (95% CI)	p	OR (95% CI)	p
Per 1 L increase	1.46 (1.12, 1.92)	.005	1.39 (1.03, 1.89)	.033	2.51 (1.43, 4.40)	.001
≦2.5 L	Reference		Reference		Reference
>2.5 L	4.25 (1.62, 11.14)	.003	5.29 (1.66, 16.85)	.004	59.90 (7.55, 475.08)	<.001

NEWS: National Early Warning Score; OR: odds ratio; CI: confidence interval.

Adjust model I adjusted for sex, age, type of heat stroke, and creatinine. Adjust model II adjusted for institution, NEWS, creatinine, troponin I, lactic acid, chlorine, coronary heart disease, use of vasoactive drugs, and length of stay in emergency department.

Table 3. Multivariate regression of normal saline infusion in emergency department for the second outcomes in patients with heat stroke.

Volume of normal saline infusion (L)	OR/β (95% CI) p
	Univariate	Adjust model I	Adjust model II
RRT	0.93 (0.54, 1.61) .793	0.70 (0.37, 1.34) .280	0.40 (0.13, 1.22) .107^a
Admitted to ICU	1.50 (1.16, 1.94) .002	1.43 (1.08, 1.91) .013	3.46 (1.58, 7.54) .002b
Length of stay in ICU	1.18 (0.70, 1.67) <.001	1.22 (0.70, 1.74) <.001	1.00 (0.44, 1.56) <.001^c
Length of stay in hospital	2.52 (1.74, 3.29) <.001	2.64 (1.85, 3.43) <.001	1.41 (0.37, 2.45) .008^d
Mortality	1.54 (1.04, 2.27) .030	2.08 (1.23, 3.53) .006	1.96 (0.72, 5.35) .189^e

NEWS: National Early Warning Score; RRT: renal replacement therapy; ICU: intensive care unit; OR: odds ratio; CI: confidence interval.

Adjust model I adjusted for sex, age, type of heat stroke, creatinine.

Adjust model II adjusted for:

^a Institution, NEWS, creatinine, lactic acid, and use of vasoactive drugs.

^b Institution, sex, age, NEWS, chlorine, creatinine, lactic acid, use of vasoactive drugs, and length of stay in emergency department.

^c Institution, type of heat stroke, age, NEWS, creatinine, lactic acid, use of vasoactive drugs, and length of stay in emergency department.

^d Institution, type of heat stroke, age, NEWS, creatinine, lactic acid, use of vasoactive drugs, and length of stay in emergency department.

^e Institution, type of heat stroke, sex, age, NEWS, chlorine, creatinine, troponin I, use of vasoactive drugs, and length of stay in emergency department.

Table 4. Multivariate regression of volume of normal saline infusion with outcomes in the genetic matching cohort^a.

Volume of normal saline infusion	Univariate		Adjust model
	OR/β (95% CI)	p Value	OR/β (95% CI)	p Value
≦2.5 L	Reference		Reference
>2.5 L
AKI	6.35 (2.66, 15.13)	<.001	7.39 (3.01, 18.11)	<.001
Admitted to ICU	3.17 (1.81, 5.54)	<.001	3.37 (1.91, 5.97)	<.001
Length of stay in ICU	2.81 (1.78, 3.85)	<.001	3.06 (2.06, 4.05)	<.001
Length of stay in hospital	3.57 (2.21, 4.94)	<.001	3.70 (2.34, 5.07)	<.001

OR: odds ratio; CI: confidence interval: ICU: intensive care unit.

Adjust model adjusted for: propensity-score.

^a The genetic matching cohort included 108 patients with volume of normal saline infusion ≦2.5 L, and 108 volume of normal saline infusion >2.5 L.

Figure 2. Smoothing spline curve of the relationship between the NS infusion volume and serum chloride levels after saline infusion. Increasing NS infusion volume was significantly associated with increasing serum chloride levels after NS infusion.

Figure 3. Smoothing spline curve of the association between serum chloride levels after saline infusion and AKI. The relationship between chloride and AKI showed a nearly S-shaped curve. The incidence of AKI increased sharply with increasing chloride when the chloride level was over 100 mmol/L.

Data availability statement

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.