Poisoning is a global problem and is a medico toxic emergency associated with high mortality rate. Though incidence of acute poisoning in Bangladesh is gradually increasing the exact scenario of poisoning is still unknown due to lack of organized poisoning database or national poisoning information center.¹¹ A 3-year study conducted in western Bangladesh at Rajshahi Medical College Hospital revealed that 4.13% of admitted patients were treated due to poisoning.¹²

In this study 86.3 % patients were less than 30 years of age group. Similarly, Dash et al showed that in India most poisoned patients (92.0%) were less than 40 years of age.¹³ Females (68.2%) were predominant in this study. Majority (38 patients, 57.6%) of the study population were married. Sobhani AR et al found in their study that 51.4% were female and 48.6% were male which is similar to this study.¹⁴ Female predominance in this study might be due to familial disharmony, poverty, failure in the examination, sexual abuse, chronic illness as well as depression leading to self-poisoning with common drugs.

Sarkar D et al. showed among pharmaceutical agents, benzodiazepines (40.0%), antipsychotics (23.0%), tricyclic antidepressants (21.0%), paracetamol (8.0%) and cetirizine (4.0%) were the most common drugs taken by patients. The most commonly used benzodiazepines (42.0% of drugs) included diazepam, bromazepam, midazolam and clonazepam.¹⁵ In our study also benzodiazepines (sedatives) were the most common group (47.0%) of drugs. This is because in Bangladesh, most medications are easily available in markets and can be sold without prescriptions. In a similar study by Nag et al. revealed that in Bangladesh, diazepam, a benzodiazepine, was the second most common cause of poisoning after OPC; which is consistent with our study.¹²

Majority patients (21 patients, 67.7%) with benzodiazepine poisoning were found to have normal ECG. Other cases of benzodiazepine overdose revealed sinus tachycardia in 7 (22.6%), prolong QT interval in 2 (6.45%) and atrial fibrillation in 1 (3.23%) patients. Eizadi-Mood N et al. showed 96.67% of poisoned patients with benzodiazepines had normal ECG, 5.25% had sinus tachycardia and 1.87% had sinus bradycardia.¹⁶ These findings are similar to our findings. In another study, Worthley Lig et al. showed that in up to 30.0% of cases with severe poisoning of benzodiazepines (particularly with respiratory acidosis) there are supraventricular and ventricular arrhythmias. In this study 1 (3.23%) patient with sedative poisoning was found to develop atrial fibrillation, which is a supraventricular arrhythmia.¹⁷

In patients with TCA (Inhibitors of fast Na+ channel) poisoning, variety of ECG findings were observed including sinus tachycardia in 11 (61.1%), prolong QT interval in 6 (33.3%), wide QRS in 5 (27.8%), atrial tachycardia in 5 (27.8%), prolong PR interval in 2 (11.1%) and tall R in aVR in 2 (11.1%) cases. A prospective study by Boehnert et al. in TCA poisoned patients demonstrated that a QRS duration less than 100 ms was an indicator of good prognosis, while those with a QRS duration over 100 ms presented with seizures in one third of cases.¹⁸ Wide QRS (>100 ms) was found in 27% of TCA poisoning patients in their study while we found similar changes in 27.8% cases of TCA poisoning (table IV). In the same study by Boehnert et al. a QRS complex over 160 ms was associated with ventricular dysrhythmias.¹⁸

Harrigan RA et al. showed that ECG changes due to TCA toxicity are attributable primarily to the Na⁺ channel blockade besides sinus tachycardia which is due to anticholinergic effects. The majority of patients with significant risk for developing cardiac or neurological toxicity will have a QRS >100 ms.¹⁹

In this study 33.3% of TCA poisoning patients had prolong QT interval (figure 2). This finding is consistent with other case reports of TCA poisoning by Vieweg at al. Even among the patients who take therapeutic doses of the tricyclic antidepressants, 20% may display mild ECG changes including prolongation of QT interval.^20,21 Other ECG changes associated with TCA overdose are sinus tachycardia, prolonged PR and QRS, ventricular arrhythmias, Brugada sign and ST segment elevation resembling ischaemic heart disease.²²

K+ efflux channel blocker overdose showed prolong QT Interval in 75.0% (3 out of 4) cases (table IV). Chlorpromazine, chlorpheniramine & quetiapine, which are K+ efflux channel blockers, were found to display prolong QT Interval in overdoses. Yates C et al. showed that amitriptyline, nortriptyline, chlorpromazine and quetiapine caused prolong QT Interval which is consistent with the findings of the study.²³ A case series by Balit et al. of 45 quetiapine overdoses revealed Prolong QT in 30.0% cases.²⁴

Prolong QT intervals are most commonly associated with TCA like amitriptyline and other antidepressants like doxepin, imipramine, and clomipramine poisoning.²⁵ In this study also TCA was the most common group that produced prolong QT interval (6 out of 11 cases of prolong QT interval) (table IV). Prolong QT interval was found with higher doses (750 mg to 1000 mg, 30-40 tablets) of amitriptyline poisoning. Poisoning with less than 30 tablets (<750 mg) of TCA mostly revealed sinus tachycardia (figure 2).

Love JN et al. showed out of 167 patients, 13 were determined to have symptomatic exposures.²⁶ First-degree heart block (>200 ms) was the most common ECG finding (10/12). But we found all patients with beta blocker overdose revealed sinus bradycardia; while only 1 (25.0%) patient with 10 tablets of atenolol induced both sinus bradycardia and prolong PR interval; this difference might be due to the lower dose of beta blockers ingested by this study patients.

Overdoses of paracetamol (500 mg tablets) were observed to display normal ECG, ST-T depression and T inversion. Similarly Weston MJ et al. showed that ECG abnormalities, especially of the ST segment or T wave and Dysrhythmias are frequent findings in paracetamol poisoning.²⁷

In this study, one patient with 6 tablets of 0.25 mg digoxin (Na-K ATPase blocker) poisoning was found to have atrial tachycardia. Christopher P et al. showed paroxysmal atrial tachycardia with variable block or accelerated junctional rhythm is highly suggestive of digitalis toxicity.²⁸

Yap G et al. showed in his study that antiarrhythmic drugs, non-sedating antihistamines, macrolide antibiotics, antifungals, antimalarials, tricyclic antidepressants, neuroleptics, and prokinetics have all been implicated in causing QT prolongation and/or torsades de pointes. In this study TCA (amitriptylines and nortriptylines), benzodiazepines, chlorpromazine & anti histamines were found to develop prolong QT Interval which is similar to the study findings by Yap G et al.²⁵

This study showed if the quantity of drug consumed by the patients is higher then ECG reveals specific abnormality. Prolong QT interval and sinus tachycardia were found in patients who took 20-25 tablets of bromazepam and diazepam (figure 1). If the quantity of TCA taken were higher (30-40 tablets/750 mg to 1000 mg), then ECG findings were abnormal like atrial tachycardia, wide QRS, prolong QT interval & prolong PR interval (figure 2). Lower doses of TCA (<30 tablets/ <750 mg) revealed mostly sinus tachycardia.

In adults, the toxic dose of TCAs varies according to the pharmacological characteristics of each one. In case of amitriptyline, low dose of 50 mg may be associated with moderate toxicity and severe toxicity occurs above 300 mg, ranging from 150 mg to 10 mg.²⁹ In our study 5 out of 18 patients of TCA overdose (>750 mg of amitriptyline) ultimately died. 2 of them had wide QRS, 2 had prolong QT & 1 had tall R in aVR. 27.8% of TCA poisoning cases revealed wide QRS. Prolong QT was found in 33.3% of TCA poisoning and 75% cases of K efflux blockers overdose. 50.0% (cases) of patients with K efflux blockers overdose died whose ECG showed prolong QT (table V).

In this study 5-10 tablets (250 mg to 500 mg) of beta blockers overdose showed sinus bradycardia in 4 (100.0%) cases and prolong PR Interval in 1 (25.0%) case. Anti-histamine produced normal ECG findings even with 20 tablets. Ten tablets of K+ efflux channel blocker (Loratadine) overdose displayed normal ECG, while 20 tablets (chlorpromazine, chlorpheniramine and quetiapine) were found to induce prolong QT Interval.

Though poisoning is a very common cause of hospital admission and mortality, no study demonstrated ECG changes in relation to drug poisoning in the context of Bangladesh. This study reflects that dangerous ECG changes like prolong QT, wide QRS and tall R in aVR may occur in higher doses of some specific group of drugs. Besides benzodiazepines, TCA is a very common drug group, which is frequently taken by the patients and associated with serious ECG changes. Mortality mostly occurred in patients whose ECG revealed prolong QT, wide QRS and tall R in aVR.

The study showed that amitriptyline & sedatives poisoning were common among young females. These drugs have their own impact and produce different types of ECG changes. ECG is a readily available tool even in remote centres. Attention to specific ECG changes of the poisoning patients can give a clue to the possible underlying drugs and also can guide immediate care. Certain ECG abnormalities also predict the prognosis of the patients. Large scale study with multi stakeholder should be under taken which will include the effect of elapsed time between the drug ingestion and the ECG changes.

We gratefully acknowledge the sincere cooperation of the patients as well as the contribution of all doctors, staffs of the department of Medicine, Dhaka Medical College Hospital.

Conflict of interest: None.