Efficacy of L-carnitine and propranolol in the management of acute theophylline toxicity

Abstract

Theophylline toxicity results in substantial morbidity and mortality particularly due to its narrow therapeutic index. The development of new treatments for acute theophylline toxicity is a point of research interest. The aim of the present work was to assess the efficacy of L-carnitine (LC) and propranolol in the management of patients with acute theophylline toxicity. The study was conducted on 60 patients with acute theophylline toxicity admitted to the Poison Control Center or Intensive Care Unit at Alexandria Main University Hospital. The studied patients were equally classified into four groups (GPs, 15 patients each): the first group was the control group who received standard treatment protocol for theophylline toxicity. The other three GPs also received standard treatment protocol for theophylline toxicity in addition. The second group (LC group) received LC with a loading dose of 100 mg/kg IV over 30–60 min (maximum 6 g) and the maintenance dose was 50 mg/kg IV every 8 h. The third group (propranolol group) received propranolol, administered slowly intravenous 0.5–1 mg over 1 min; it may be repeated if necessary up to a total maximum dose of 0.1 mg/kg. The fourth group (propranolol and LC) received both IV propranolol and LC in the same doses as previous. Treatment with LC alone or in combination with propranolol resulted in a significant improvement of both clinical and laboratory findings. Although combined therapy yields the best results and outcome, LC can serve as an antidote for acute theophylline toxicity if there is any contraindication to propranolol administration.

Introduction

Despite the marked decrease in the use of theophylline (a naturally occurring methylxanthine) as a treatment of asthma because of its narrow therapeutic window, yet theophylline intoxication episodes still occur resulting in substantial morbidity and mortality [1]. Theophylline was the second most frequent drug involved in poisoned cases received in the poison control center of Ain Shams University Hospital (PCC, ASUH) through the year 2011 [2]. Theophylline also was the commonest offending drug that caused cardiovascular presentation in the poisoned cases admitted to PCC, ASUH through the year 2013 [3].

Theophylline has a poor therapeutic effect in cases of acute airway spasm, but its continued prescription is mainly for asthma prophylaxis due to its bronchodilator, anti-inflammatory as well as immunomodulatory effects together with its pulmonary vasodilation effect, stimulation of mucociliary clearance and inhibition of mast cell mediator response [4]. It is also used in the treatment of neonatal apnea because of its stimulatory effect on the medullary respiratory center [5].

Minor toxic manifestations include vomiting, tremor or cardiac disturbances without the occurrence of hemodynamic instability, whereas major life-threatening toxic manifestations involved seizures or arrhythmias, resistant to standard therapeutic interventions and combined with hemodynamic compromise [6].

Mechanisms by which xanthine causing toxicities were competitive antagonism of adenosine receptors, inhibition of prostaglandin, enhancing the release of endogenous norepinephrine, and consequent adrenergic receptor stimulation and inhibition of cyclic nucleotide phosphodiesterase (PDE) thus preventing the breakdown of cyclic adenosine monophosphate (cAMP) and guanosine monophosphate (cGMP), translocation of intracellular calcium ions. Moreover, xanthine stimulates β2 adrenergic receptors to increase cAMP with resultant bronchodilatation [4, 7, 8].

It was reported that theophylline-induced tachycardia and seizures were accompanied by alterations in the levels of oxidative stress biochemical markers. This shows that oxidative stress may be involved in theophylline-induced toxicity and that antioxidants may have a role in management [9, 10].

Propranolol is a nonselective beta-blocker. It reverses peripheral vasodilation, hypotension via β₂-adrenergic receptors blockage, and reverses tachycardia via β₁-adrenergic receptors blockage [11].

L-Carnitine (LC) is an effective antioxidant agent, has an important role in normal mitochondrial oxidation of fatty acids. It can protect human endothelial cells against lipid peroxidation and decreases xanthine oxidase activity [12].

Actually, acute theophylline toxicity constitutes a common toxicity especially in developing countries because of its cheap price and availability of over-the-counter preparations, 10-fold dosing error [13], Intake of sustained-release formulations and absence of specific antidote and presence of drug interactions affecting the drug metabolism and clearance aggravate the problem. Hence, the need developed for recognizing the efficacy of some drugs like LC and propranolol in the management of acute theophylline toxicity, which was the target of the present work.

Subjects and Methods

The study was conducted on 60 patients with acute theophylline toxicity admitted to the Poison Control Center or Intensive Care Unit at Alexandria Main University Hospital.

The patients were enrolled in the study in concordance to the following inclusion criteria: age ranged from 16 to 70 years [4, 14, 15]. Patients admitted within 24 h from exposure to the poison without any prior treatment before admission.

Patients were excluded if they had any of the following conditions: contraindication to propranolol [16], smokers, patients with severe hypoalbuminemia (albumin < 2 g/dl) including cases of chronic malnutrition, hemorrhage, protein-losing nephropathy, enteropathy, hepatic failure [17], patients having malignancy, diabetes mellitus, or cardiac diseases, in addition to cases of coingestion.

The current study was carried out after approval of the Ethics Committee of Faculty of Medicine, Alexandria University (IRB number: 00007555, approval serial number: 0105608). Informed consent was obtained from every patient included in the study or from his family, explaining the aim and the procedure of the research. Complete confidentiality was ensured all through the study procedures.

Diagnosis of acute theophylline toxicity was performed by history-taking or checking the recently used tablets, or by characteristic symptoms and signs of toxicity as well as by serum theophylline concentration.

All patients were subjected to thorough history taking, complete physical examination. Electrocardiography (ECG) was performed and laboratory investigations were measured for every patient both on admission and 24 h after the start of treatment, including serum theophylline level, arterial blood gases, and serum oxidative stress markers: glutathione (GSH) and malondialdehyde (MDA).

The standard treatment protocol for theophylline toxicity was given to all patients. It included maintenance of airway and breathing and stabilization of circulation by giving IV fluids (crystalloids) to treat hypotension and use vasopressors if needed with the management of arrhythmia. Decontamination using activated charcoal was performed if vomiting stopped early. Gastrointestinal tract (GIT) disturbance was managed by antiemetic (ondansetron) and H₂ blocker (ranitidine). Treatment of seizures and other complications was done. Close monitoring of serum potassium values with potassium replacement via IV infusion. Elimination can be enhanced by hemodialysis if indicated [11, 18–20].

The patients were divided into four equal groups (GPs) 15 patients each) as follows: the first group was the control group who received standard treatment protocol for theophylline toxicity. The other three GPs also received standard treatment protocol for theophylline toxicity in addition. The second group (LC group) received LC with a loading dose of 100 mg/kg IV over 30–60 min (maximum 6 g) and the maintenance dose was 50 mg/kg IV every 8 h [21]. The third group (propranolol group) received beta-blocker (propranolol) administered slowly intravenous 0.5–1 mg over 1 min; it may be repeated if necessary up to a total maximum dose of 0.1 mg/kg [22]. The fourth group (propranolol and LC) received both IV propranolol and LC in the same doses as previous.

The outcome was assessed according to the length of hospitalization and the need for intensive care unit (ICU) admission and categorized into three categories: recurrence of cardiovascular system (CVS) symptoms, improvement/recovery or death.

Statistical analysis of the data

The sample size was calculated using a one-way analysis of variance (ANOVA) test (PASS program version 20).

Data were analyzed using IBM SPSS software package version 20.0. Qualitative data were described using numbers and percentages. Categorical data were evaluated by the chi-square test and Monte Carlo correction. The Kolmogorov–Smirnov test was used to verify the normality of distribution. For normally distributed quantitative variables, a paired t-test was used to compare between two periods, and F-test (ANOVA) to compare between more than two GPs. While for abnormally distributed quantitative variables, Wilcoxon-signed rank test was used to compare between two periods and Kruskal–Wallis test to compare between more than two studied GPs. The significance of the obtained results was examined at the 5% level.

Results

The study enrolled 60 patients; 53 patients were females (88.3%), whereas seven patients were males (11.7%) with a sex ratio of 7.57:1. The age of the patients ranged from 16 to 52 years, with a mean age of 23.20 ± 8.12 years. The majority of cases (80%) took theophylline to commit suicide and 20% of cases were accidental poisoning. No statistically significant differences were detected among the four studied GPs regarding age, sex, marital status, education, and circumstances of poisoning (Table 1).

Table 1.

Comparison between the four studied GPs according to personal history and circumstances of poisoning

Personal history	GP I (n = 15)		GP II (n = 15)		GP III (n = 15)		GP IV (n = 15)		Total		Test of significance	P *
	No.	%	No.	%	No.	%	No.	%	No.	%
Age (years)
≤30	12	80.0	13	86.7	14	93.3	14	93.3	53	88.33	χ 2 7.596	MC P = 0.967
>30	3	20.0	2	13.3	1	6.7	1	6.7	7	11.67
Min.–Max.	16.0–52.0		16.0–40.0		16.0–40.0		17.0–40.0		16.0–52.0		H = 1.143	0.767
Mean ± SD	24.73 ± 11.37		22.53 ± 7.70		22.93 ± 6.41		22.60 ± 6.67		23.20 ± 8.12
Gender
Male	1	6.7	2	13.3	3	20.0	1	6.7	7	11.7	χ 2 = 1.729	MC P = 0.822
Female	14	93.3	13	86.7	12	80.0	14	93.3	53	88.3
Marital status
Single	10	66.7	10	66.7	10	66.7	12	80.0	42	70.0	χ 2 = 2.102	MC P = 0.972
Married	4	26.7	4	26.7	4	26.7	3	20.0	15	25.0
Divorced	1	6.7	1	6.7	1	6.7	0	0.0	3	5.0
Education
Illiterate	3	20.0	4	26.7	4	26.7	4	26.7	15	25.0	χ 2 = 0.909	MC P = 1.000
Pre-university	8	53.3	8	53.3	7	46.7	8	53.3	31	51.7
University	4	26.7	3	20.0	4	26.7	3	20.0	14	23.3
Circumstances of poisoning
Suicidal	11	73.3	12	80.0	14	93.3	11	73.3	48	80.0	χ 2 = 2.648	MC P = 0.489
Accidental	4	26.7	3	20.0	1	6.7	4	26.7	12	20.0

χ ², chi-square test; MC, Monte Carlo; H, H for Kruskal–Wallis test; GP I, control group; GP III, propranolol group; GP II, LC group; GP IV, propranolol and LC.

^* P-value for comparison between the four studied GPs.

There was no statistically significant difference among the four GPs regarding the time of starting treatment after exposure. The serum theophylline level was significantly lower after 24 h of admission than that on admission in all GPs. The highest % change or decrease in serum theophylline level was detected in GPs received LC (GP IV, % change = 86% and GP II, % change 84.9%) (Table 2).

Table 2.

Comparison between the four studied GPs according to time of starting treatment and serum theophylline level

	GP I (n = 15)	GP II (n = 15)	GP III (n = 15)	GP IV (n = 15)	H	P
Time of starting treatment after exposure (h)
Min.–Max.	1.50–14.0	2.0–10.0	2.50–12.0	3.0–14.0	H = 1.324	0.723
Mean ± SD	5.93 ± 3.76	5.53 ± 2.70	5.60 ± 2.84	6.60 ± 3.29
Serum theophylline level (mg/L)
On admission
Min.–Max.	21.60–88.90	26.60–97.0	34.40–88.30	35.20–90.50	3.467	0.325
Mean ± SD	57.19 ± 21.23	55.58 ± 24.19	62.19 ± 18.24	69.05 ± 15.06
After 24 h
Min.–Max.	4.90–29.0	0.60–17.0	8.20–26.30	3.40–21.20	16.862*	0.001*
Mean ± SD.	16.39 ± 8.50	8.47 ± 4.86	17.09 ± 5.68	9.79 ± 4.05
Pcontrol		0.006*	0.530	0.019*
Significance between GPs		P 1 = 0.001*, P2 = 0.683, and P3 = 0.003*
P0	0.001*	0.001*	0.001*	0.001*
% Change	↓72.0	↓84.9	↓72.2	↓86.0

Pairwise comparison between each two GPs was carried out using post hoc test (Dunn’s for multiple comparisons test). H, H for Kruskal–Wallis test; χ², chi-square test; MC, Monte Carlo; P, P-value for comparison between the four studied GPs; P₀, comparison between on admission and after 24 h in each GP; P_control, P-value for comparison between GP I and each other GP; P₁, P value for comparison between GP II and GP III; P₂, P value for comparison between GP II and GP IV; P₃, P value for comparison between GP III and GP IV.

^*Statistically significant at P ≤ 0.05.

A statistically significant difference was not found among the four GPs regarding the measured serum theophylline level on admission and the difference was developed among them only after 24 h of the start of therapeutic intervention. A statistically significant difference detected only between GP II and either control or GP III and between GP IV and either control or GP III (between any of the two GPs received carnitine and other two GPs; Table 2).

The heart rate (HR) significantly decreased after passing 24 h of therapeutic intervention than that on admission in all GPs. The highest % change or decrease in HR was detected in the GP received combined therapy of propranolol and LC (GP IV; 32.36%) and it was the only GP that had a statistically significant difference with the control GP after 24 h of applying the treatment (Table 3).

Table 3.

Comparison between the four studied GPs according to HR and mean arterial blood pressure

		GP I (n = 15)	GP II (n = 15)	GP III (n = 15)	GP IV (n = 15)	F	P
HR (beat/min)	On admission
	Min.–Max.	95.0–190.0	100.0–170.0	125.0–150.0	100.0–166.0	1.191	0.322
	Mean ± SD.	128.3 ± 26.14	130.1 ± 17.27	139.0 ± 8.49	137.1 ± 17.82
	On admission
	Min.–Max.	80.0–135.0	75.0–125.0	75.0–110.0	70.0–110.0	2.781*	0.049*
	Mean ± SD	105.0 ± 15.47	97.73 ± 14.22	94.13 ± 9.48	92.33 ± 12.08
	Pcontrol		0.427	0.113	0.048*
	Significance between GPs		P 1 = 0.873, P2 = 0.669, and P3 = 0.981
	P0	<0.001*	<0.001*	<0.001*	<0.001*
	% change	↓17.06	↓24.68	↓32.26	↓32.36
Mean BP (mmhg)	On admission
	Min.–Max.	60.0–113.0	60.0–97.0	40.0–103.0	53.0–100.0	0.655	0.583
	Mean ± SD.	76.07 ± 14.61	69.33 ± 9.41	70.80 ± 17.01	73.33 ± 14.53
	After 24 h
	Min.–Max.	70.0–103.0	83.0–100.0	70.0–108.0	80.0–108.0	3.777*	0.015*
	Mean ± SD.	83.67 ± 8.94	90.40 ± 5.38	88.13 ± 10.91	94.07 ± 8.54
	Pcontrol		0.157	0.498	0.009*
	Significance between GPs		P 1 = 0.890, P2 = 0.656, and P3 = 0.251
	P0	0.011*	<0.001*	<0.001*	<0.001*
	% Change	↑11.9	↑32.1	↑28.8	↑31.6

Pairwise comparison between each two GPs was carried out using post hoc test (Tukey). F, F for ANOVA test; P, P-value for comparison between the four studied GPs; P₀, comparison between on admission and after 24 h in each GP; P_control, P value for comparison between GP I and each other GP; P₁, P value for comparison between GP II and GP III; P₂, P value for comparison between GP II and GP IV; P₃, P value for comparison between GP III and GP IV.

^*Statistically significant at P ≤ 0.05.

A statistically significant difference developed among the four GPs regarding ECG only after passing 24 h after the treatment, where 73.3% of GP I patients presented with different types of arrhythmias, whereas the least percentage of arrhythmia was detected in GPs received propranolol; GP III and IV (13.3% each). The highest % change or decrease in sinus arrhythmia occurred with GP IV (73.4%) (Figs 1 and 2).

Figure 1.

Comparison between the four studied GPs according to heart rhythm on admission (GP I, Control group; GP III, propranolol group; GP II, LC group; GP IV, propranolol and LC).

Figure 2.

Comparison between the four studied GPs according to heart rhythm after 24 h.

As regards to mean arterial pressure (MAP), it was significantly higher after 24 h of admission than that on admission in all GPs. GPs received LC showed the highest % increase in MAP (32.1% in GP II, 31.6% in GP IV)). A statistically significant difference developed among the four GPs after 24 h. GP IV was the only GP that had a significantly higher MAP than that of the control GP (Table 3).

The respiratory rate (RR) on admission was high in all four GPs. RR significantly decreased after 24 h in all GPs except the control GP. The highest % decrease in RR presented in (GP IV; 37%) (Table 4).

Table 4.

Comparison between the four studied GPs according to RR

RR	GP I (n = 15)	GP II (n = 15)	GP III (n = 15)	GP IV (n = 15)	F	P
On admission
Min.–Max.	25.0–40.0	25.0–40.0	25.0–40.0	25.0–50.0	1.104	0.355
Mean ± SD.	32.13 ± 4.24	31.93 ± 5.16	30.47 ± 4.34	34.0 ± 7.12
After 24 h
Min.–Max.	20.0–45.0	19.0–30.0	16.0–27.0	16.0–28.0	11.352*	<0.001*
Mean ± SD	30.0 ± 7.19	23.53 ± 3.56	21.73 ± 3.17	21.07 ± 3.67
Pcontrol		0.002*	<0.001*	<0.001*
Significance between GPs		P 2 = 0.720, P3 = 0.480, and P4 = 0.980
P0	0.247	<0.001*	<0.001*	<0.001*
% Change	↓6.1	↓26.0	↓27.9	↓37.0

Pairwise comparison between each two GPs was carried out using post hoc test (Tukey). F, F for ANOVA test; P, P-value for comparison between the four studied GPs; P₀, comparison between on admission and after 24 h in each GP.

^*Statistically significant at P ≤ 0.05.

There were statistical significant differences among the four studied GPs regarding the mean levels of MDA and GSH after 24 h of admission The control and propranolol GPs had a high % of increase MDA levels after 24 h of admission (167.3% and 153.7% respectively), while both GPs received LC showed % decrease in MDA levels (60.5% in GP II, 51.6% in GP IV) (Fig. 3).

Figure 3.

Comparison between the four studied GPs according to mean MDA level (nmol/ml)

On the other hand, all GPs had significantly higher mean GSH levels after 24 h than those on admission. GP II that received LC only had the highest % of increased GSH levels after 24 h of admission (147.7%) (Fig. 4).

Figure 4.

Comparison between the four studied GPs according to mean GSH level (mg/dl)

The mean PH decreased after 24 h of admission in all GPs. The lowest % decrease in PH was in the two GPs received carnitine (GP II; ↓0.33, GP IV; ↓0.03). On the other hand, the mean P_aCO₂ increased after 24 h of admission in all GPs, with the highest % increase in P_aCO₂ recorded in the same two GPs received carnitine (GP II; ↑30.67, GP IV; ↑23.47) (Figs 5 and 6).

Figure 5.

Comparison between the four studied GPs according to mean PH

Figure 6.

Comparison between the four studied GPs according to mean P_aCO₂ level (mmHg)

The mean serum HCO₃ showed decreased levels in all GPs on time of admission, which improved after 24 h in all GPs with statistically significant differences between its mean levels in the two measuring times in only two GPs that received LC (GP II and GP IV). The highest % increase in HCO₃ levels was recorded in the same two GPs received carnitine (GP II; ↑23.28, GP IV; ↑34.72) (Table 5).

Table 5.

Comparison between the four studied GPs according to mean HCO₃ levels (mmHg)

HCO3	GP I (n = 15)	GP II (n = 15)	GP III (n = 15)	GP IV (n = 15)	F	P
On admission
Min.–Max.	13.50–23.70	8.90–24.60	13.50–23.80	11.30–27.60	1.252	0.300
Mean ± SD	19.82 ± 2.38	18.93 ± 3.69	19.10 ± 2.89	17.44 ± 4.50
After 24 h
Min.–Max.	15.0–22.90	19.70–24.60	16.70–23.40	19.90–25.0	9.646*	<0.001*
Mean ± SD	19.63 ± 2.23	22.35 ± 1.70	19.96 ± 1.81	22.34 ± 1.55
Pcontrol		0.001*	0.962	0.001*
Significance between GPs		P 1 = 0.004*, P2 = 1.000, and P3 = 0.004*
P0	0.790	<0.001*	0.051	<0.001*
% Change	↑0.12	↑23.28	↑5.70	↑34.72

Pairwise comparison between each two GPs was carried out using post hoc test (Tukey). F, F for ANOVA test; P, P-value for comparison between the four studied GPs; P₀, comparison between on admission and after 24 h in each GP; P_control, P value for comparison between GP I and each other GP; P₁, P value for comparison between GP II and GP III; P₂, P value for comparison between GP II and GP IV; P₃, P value for comparison between GP III and GP IV.

^*Statistically significant at P ≤ 0.05.

There were highly statistically significant differences among the four studied GPs regarding both the total amount of fluid and antiemetic taken within 24 h. The control GP had the highest mean total fluid taken, whereas the lowest mean fluid taken was in GPs received LC (GP II and IV). GP IV had the lowest mean among all studied GPs regarding both lines of treatment (Table 6).

Table 6.

Comparison between the four studied GPs according to management

Management	GP I (n = 15)		GP II (n = 15)		GP III (n = 15)		GP IV (n = 15)		Test of significance	P
	No.	%	No.	%	No.	%	No.	%
Total IV fluids in 24 h
Min.–Max.	2.0–3.50		1.50–3.50		1.50–3.50		1.50–3.0		F = 7.026*	<0.001*
Mean ± SD	2.97 ± 0.52		2.27 ± 0.53		2.70 ± 0.46		2.23 ± 0.56
Pcontrol			0.003*		0.498		0.002*
Significance between GPs			P 1 = 0.112, P2 = 0.998, and P3 = 0.076
Anti-emetic (ondansetron 4 mg)
Min.–Max.	8.0–24.0		4.0–12.0		4.0–24.0		1.50–3.0		H = 43.662*	<0.001*
Mean ± SD	14.67 ± 4.19		8.53 ± 2.56		12.8 ± 4.83		2.23 ± 0.56
Pcontrol			0.003*		0.390		<0.001*
Significance between GPs			P 1 = 0.039*, P2 = 0.002*, and P3 < 0.001*

Pairwise comparison between each two GPs was carried out using post hoc test (Tukey). χ², chi square test; MC, Monte Carlo; H, H for Kruskal–Wallis test; F, F for ANOVA test.

Only two cases in the control GP needed ICU admission and one of them needed MV. The same occurred with GP IV. However, the duration of stay in ICU was longer in the control GP than that in GP IV, reaching 5 days in the control GP versus 2 days maximally in GP IV. There were statistically significant differences among them regarding the days stayed in the poison center. The control GP had the highest mean days of hospital stay and GP IV had the lowest mean days of stay. There were also statistically significant differences among the four GPs regarding the outcome, where recurrence of symptoms occurred in 66.7% of control cases and the highest percentage of improved cases was in GP IV (86.7%) (Table 7).

Table 7.

Comparison between the four studied GPs according to outcome

Outcome	GP I (n = 15)		GP II (n = 15)		GP III (n = 15)		GP IV (n = 15)		Test of significance	P
	No.	%	No.	%	No.	%	No.	%
Length of hospital stay (days)
Days in poison centre	(n = 15)		(n = 15)		(n = 15)		(n = 15)
Min.–Max.	1.0–4.0		1.0–3.0		1.0–3.0		1.0–3.0		F = 2.765*	0.049*
Mean ± SD	2.40 ± 0.74		2.13 ± 0.74		2.27 ± 0.59		1.73 ± 0.59
Days in ICU	(n = 2)		(n = 0)		(n = 0)		(n = 2)
Min.–Max.	2.0–5.0						1.0–2.0		#	#
Mean ± SD.	3.50 ± 2.12		–		–		1.50 ± 0.71
Outcome
Recurrence of symptoms	10	66.7	5	33.3	5	33.3	2	13.3	χ 2 = 14.013*	MC P = 0.023*
Improvement/recovery	5	33.3	10	67.7	10	67.7	13	86.7
Died	0	0.0	0	0.0	0	0.0	0	0.0
Pcontrol			0.068		0.068		0.003*
Significance between GPs			P 1 = 1.000, FEP2 = 0.390, and FEP3 = 0.390

Pairwise comparison between each two GPs was carried out using post hoc test (Dunn’s for multiple comparisons test). χ², Chi square test; MC, Monte Carlo; H, H for Kruskal–Wallis test; P, P value for comparison between the four studied GPs.

^*Statistically significant at P ≤ 0.05.

^#Excluded from the comparison due to small number of cases.

Discussion

The study enrolled 60 patients with a female to male sex ratio of 7.57:1. The majority of cases (80%) took theophylline to commit suicide.

The age of the patients ranged from 16 to 52 years, as age is one of the important factors affecting theophylline clearance. Theophylline clearance is progressively increased from birth reaching the maximum clearance in young children and stabilizes in the mid-teens then decreases again in the elderly [4, 15]. In the present work, the majority of the intoxicated patients (88.33%) were in the age GP ≤ 30 years as adolescents and young adults are more prone to mood changes and attempted suicide.

No statistically significant difference was found among the four GPs regarding serum theophylline level on admission. The highest % change or decrease in serum theophylline level after 24 h of admission was detected in GPs received LC (GP IV, % change = 86% and GP II, % change = 84.9%). This may indicate the role of LC in theophylline clearance similar to enhancing clearance of valproic acid and so, was used as an antidote for it [23].

The highest % change or decrease in HR after 24 h of admission was detected in the GP received combined therapy (GP IV; 32.36%) and it was the only GP that had a statistically significant difference with the control GP after 24 h of applying the treatment.

β-Blockers have a negative chronotropic effect through blocking β1 receptors [24]. Moreover, a significant negative chronotropic effect was detected by Brooks et al. [25], reporting a reduction of HR from 124 to 86 beats/min after 1 h of intravenous LC injections in their animal studies.

After passing 24 h of the treatment, the least percentage of detected arrhythmia was in GPs received propranolol; GP III and IV (13.3% each). The highest % change or decrease in sinus arrhythmia occurred with GP IV (73.4%). β-Blocker (propranolol) is preferred in the treatment of arrhythmia of theophylline than using Ca channel blocker as the latter can worsen the state of hypotension.

Freeman and Crapo [26] and McCord et al. [27] stated that increased catecholamine stimulates the generation of free radicals via the increase in mitochondrial respiration and auto-oxidation. These free radicals negatively affect myocardial metabolism and function, inducing arrhythmias [28, 29]. This explains the role of an antioxidant in the treatment of arrhythmia.

Colbridge [4] reported that propranolol could treat supraventricular arrhythmia and Imai et al. [30] stated that carnitine significantly decreased ventricular arrhythmias even ventricular fibrillation.

As regards to MAP, GPs received LC showed the highest % increase in MAP (32.1% in GP II, 31.6% in GP IV)). A statistically significant difference developed among the four GPs after 24 h. GP IV was the only one that had a significantly higher MAP than that of the control GP.

The hypotension in theophylline is common and multifactorial and resulted from severe protracted vomiting, β2 stimulated peripheral vasodilatation and inhibition of PDE with increased cAMP causing vascular smooth muscle relaxation. In addition, theophylline stimulates β1 receptors in myocytes producing a positive chronotropic effect, which causes a decrease in time for diastolic filling aggravating the hypotensive state.

β-Blockers can treat hypotension by producing a negative chronotropic effect through blocking β1 receptors, slowing the HR, prolonging the diastolic filling time, thus increases stroke volume and improving cardiac output [24].

Propranolol (nonselective β-blocker) is superior to esmolol (selective β1 blocker; cardioselective) as it has a longer half-life (T_1/2) than that of esmolol (only 9 min), and propranolol can reverse metabolic disturbances induced by theophylline as they are β2 mediated [4, 31]. Moreover, propranolol can treat hypotension by counteracting the relaxing effect of stimulated β2 receptors on peripheral vessels as an additional mechanism [31]. However, esmolol is preferred to treat theophylline-induced hypotension in asthmatic patients to avoid bronchospasm induced by β2 blocking effect of propranolol.

Suzuki et al. [32] found that intravenous carnitine caused a 60% increase in coronary blood flow which reached over 100% increase in the study of Brooks et al. [25] with dramatic improvements in left ventricular work, stroke volume, cardiac output, and other parameters of contractility.

Gurlek et al. [33] stated that the addition of LC to the conventional therapy of ischemic cardiomyopathy was more significant than treatment with conventional treatment only as LC produced improvement in left ventricular ejection fraction.

Carnitine maintains energy (ATP) production by facilitation of the transfer of long-chain fatty acid into mitochondria (increases fat oxidation) and regulation of carbohydrate metabolism through increasing glucose oxidation. Cardiomyocytes can use fat for energy by carnitine. Carnitine also enhances lactate extraction from cardiomyocytes, thus improves oxygen consumption and exercise tolerance and many cardiac parameters and delays the onset of fatigue. So, it was indicated in various cardiac disorders. Moreover, it acts as cardioprotective by increasing the resistance of cardiac cell membranes to reactive oxygen species, by decreasing levels of toxic coenzyme A derivatives [34, 35]. Fujiwara et al. [36] reported that LC increased coronary blood flow during exercise.

The RR on admission was high in all four GPs by stimulatory action of theophylline on the respiratory center, causing hyperventilation, and respiratory alkalosis. RR was significantly decreased after 24 h in all GPs except the control GP. The highest % decrease in RR was encountered in GP IV (37%), this confirms the superiority of combined therapy in GP IV than other GPs.

Bernard [37] reported that respiratory alkalosis was the commonest acid–base disturbance encountered in acute theophylline overdose. Greene et al. [20] stated that respiratory alkalosis is common in awake patients, which coincides with the present study where 93.3% of patients were fully conscious (n = 56).

There were statistical significant differences among the four studied GPs regarding the mean levels of MDA and GSH after 24 h of admission The control and propranolol GPs had a high % of increase MDA levels after 24 h of admission (167.3% and 153.7%, respectively), while both GPs received LC showed % decrease in MDA levels (60.5% in GP II, 51.6% in GP IV). On the other hand, all GPs had significantly higher mean GSH levels after 24 h than those on admission. GP II that received LC only had the highest % of increase GSH levels after 24 h of admission (147.7%).

Oxidative stress or imbalance of redox status has been implicated in the toxicity induced by aminophylline including seizures [38]. Gulati et al. [9] reported that seizures, as well as mortality, were declined by antioxidants use (melatonin, N-acetylcysteine) and by nitric oxide (NO) synthase inhibitors. They detected high levels of MDA and NO metabolites in the brain of mice with aminophylline-induced seizures, and these metabolites decreased by antioxidant pretreatment. This was confirmed by reduction of seizures as well as mortality by 10% in case of pretreatment with LC in the study of Roy et al. [39], where the tested efficacy of LC in correction of imbalance of redox status was manifested by the significant suppression of MDA and significantly improved superoxide dismutase activity in treated rats’ brains than control GP.

Carnitine is biosynthesized from the amino acids methionine and lysine. LCs are the biologically active enantiomer having antioxidant property, it protects against oxidative stress induced at the myocardial and endothelial cell level [40]. LC counteracts oxidative stress indirectly by augmenting the activity of cellular scavengers as GSH and vitamin E or by acting as a metal chelator [41] and directly through inhibition of arachidonic acid (AA)-induced nicotinamide adenine dinucleotide phosphate oxidase activation [42].

The mean PH decreased after 24 h of admission in all GPs. The lowest % decrease in PH was in the two GPs received carnitine (GP II; ↓0.33, GP IV; ↓0.03). On the other hand, the mean P_aCO₂ increased after 24 h of admission in all GPs, with the highest % increase in P_aCO₂ recorded in the same two GPs received carnitine (GP II; ↑30.67, GP IV; ↑23.47).

The study revealed that the highest % increase in HCO₃ levels was recorded in GP IV; ↑34.72). Colbridge [4] reported that propranolol can reverse the β2-mediated metabolic disturbances induced by theophylline such as hypokalemia, hypophosphatemia, hypophosphatemia, hyperglycemia, and acidosis.

The study revealed that GP IV had the lowest mean among all studied GPs regarding the total amount of fluid and antiemetic taken within 24 h. Methylxanthines induce vomiting by multiple mechanisms; through a direct effect on medullary vomiting center, or by increasing the gastric acid and pepsin secretion causing esophagitis and gastritis, they also cause smooth muscle relaxation including relaxation of the lower esophageal sphincter [43].

It was found that LC showed a decrease in C-reactive protein, i.e. acts as an anti-inflammatory [44]. Moreover, LC strengthens the gastric mucosal barrier, thus prevents the occurrence of mucosal lesions. It acts by counteracting the effect of lipid peroxidation, hence it prevents the decrease in prostaglandin E₂ (PGE2) content of gastric mucosa, which is a potent gastroprotective [45].

The duration of stay in ICU was longer in the control GP than that in GP IV as it reached 5 days in the control GP versus 2 days maximally in GP IV. The control GP had the highest mean days of hospital stay and GP IV had the lowest mean days of stay. There were also statistically significant differences among the four GPs regarding the outcome, where recurrence of symptoms occurred in 66.7% of control cases and the highest percentage of improved cases was in GP IV (86.7%).

Oami et al. [46] reported that nearly one-fourth of patients admitted to ICU had LC deficiency and could benefit from LC supplementation. Carnitine depletion occurred in ICU admission as the patients are usually nourished through parenteral feeding, which lacks sufficient LC content and also LC may be depleted in those patients because of the metabolic load on mitochondrial function in critical illness.

LC can serve as an antidote for theophylline. Firstly, as being an antioxidant, it counteracts the oxidative stress mechanism involved in theophylline toxicity, secondly because of its safety and tolerability. Arsenian [47] recommended LC in his study as the majority of clinical studies with carnitine did not report any side effects or major clinical toxicity.

Conclusion

Early pretreatment with combined therapy; GP IV (propranolol and L-carnitine) proved its superiority over the other studied GPs in the amelioration of both clinical and laboratory manifestations of acute theophylline toxicity. After 24 h of therapeutic intervention, GP IV showed the highest % decrease in serum theophylline level, RR, HR, and sinus arrhythmia with the highest % increase in serum HCO₃ level. Moreover, GP IV had the lowest mean value of the total amount of IV fluid intake, antiemetic needed and hospital stay period. Finally, GP IV showed the best outcome (i.e. the highest % of improved or recovered cases and the lowest % of recurrence of symptoms).

Regarding the oxidative stress markers, GP II showed the best results among the four studied GPs followed by GP IV where GP II had the highest % decrease in MDA level together with the highest % increase in GSH level. Efficacy of L-carnitine alone or in combination with propranolol proved the role of oxidative stress mechanism in theophylline toxicity.

Recommendations

The study recommends modification of the existing regimen for the treatment of acute theophylline toxicity by the addition of propranolol and LC. Pretreatment with both drugs is beneficial especially before the development of manifestations. The study also recommends pretreatment with LC alone if there is any contraindication to propranolol administration as showed high efficacy also if used alone.

Funding

There are no funding sources.

Conflict of Interest Statement

There are no conflicts of interest to declare.