QJM Advance Access originally published online on April 5, 2007
QJM 2007 100(5):271-276; doi:10.1093/qjmed/hcm017
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pattern of lithium exposure predicts poisoning severity: evaluation of referrals to a regional poisons unit
From the Scottish Poisons Information Bureau, Royal Infirmary of Edinburgh, Edinburgh, UK
Address correspondence to Dr W.S. Waring, Scottish Poisons Information Bureau, Royal Infirmary of Edinburgh, 51 Little France Crescent, Edinburgh EH16 4SA. email: s.waring{at}ed.ac.uk
Received 22 November 2006 and in revised form 28 December 2006
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Background: Lithium toxicity may result in severe clinical features. There is on-going uncertainty about the significance of serum lithium concentrations in patients with lithium toxicity.
Aim: To examine potential relationships between stated quantity of lithium ingested, serum lithium concentrations, and poisoning severity among patients referred to a regional poisons centre.
Methods: Prospective evaluation of enquiries to the Scottish Poisons Information Bureau about lithium toxicity between 2000–2005 inclusive.
Results: There were 172 enquiries, relating to acute ingestion (n = 101), acute-on-therapeutic ingestion (n = 38), or chronic poisoning (n = 33). Poisoning severity was moderate or severe in 9.9%, 26.3% (p < 0.05 vs. acute) and 54.5% (p < 0.005 vs. acute) of each group, respectively. Median (IQR) serum lithium concentrations in each group were: 2.4 (1.7–3.3) mmol/l, 2.1 (1.4–3.8) mmol/l, and 2.3 (1.9–3.3) mmol/l, respectively. The median stated quantities ingested in acute and acute-on-therapeutic lithium exposure were 5000 mg (2000–11 050 mg) and 4000 mg (2400–8820 mg), respectively.
Discussion: Patients with acute-on-therapeutic and chronic poisoning are at greatest risk of severe toxicity. These differences cannot be explained by either the quantity of lithium ingested or serum lithium concentration alone.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Lithium has been used in the treatment of mania and depression for more than 50 years. Serum concentrations should normally be maintained between 0.5–0.8 mmol/l (mEq/l); the prevalence of adverse effects increases significantly at concentrations only slightly higher than these.1 Features of lithium toxicity characteristically include drowsiness, slurred speech, psychomotor slowing, polyneuropathy, and impaired memory and, in severe cases, seizures, coma and death.2–4 Other recognized effects include atrioventricular delay, arrhythmia, and acute renal failure.5–7 Clinical manifestations of lithium toxicity effects may lag behind changes in serum lithium concentrations, due at least in part to delayed distribution into tissues.8 Lithium poisoning is associated with distinctive histological abnormalities in cerebral tissue,9–12 and there may be persistent neurological sequelae despite recovery from the acute effects.13–18
Three distinct patterns of lithium toxicity are recognized: ‘acute’ in patients not previously receiving lithium, ‘acute-on-therapeutic’ due to acute ingestion in the setting of current lithium treatment, and ‘chronic’, which arises insidiously due to lithium accumulation. Each has a characteristic pattern of severity and duration of toxicity, due to differences in the unique pharmacokinetic properties that determine tissue lithium concentrations. In acute toxicity, serum lithium concentrations fall rapidly due to distribution throughout tissues, whereas patients with acute-on-therapeutic and chronic toxicity have baseline tissue saturation to some extent. Maintenance lithium treatment is associated with a progressive decline in renal function and serum lithium half-life is prolonged. Therefore, patients with acute-on-therapeutic or chronic toxicity may be exposed to higher lithium concentrations for a longer duration than patients with acute toxicity.19–21 This results in corresponding differences in poisoning severity between different patterns of toxicity.
Serum lithium concentrations are often used to guide clinical management of patients who present to hospital with lithium toxicity, for example in determining the need for haemodialysis. Despite this, serum lithium concentrations correlate with poisoning severity only in some, but not all studies of patients with chronic lithium toxicity.22,23 Indeed, there is no apparent correlation whatsoever between serum lithium concentrations and severity in acute or acute-on-therapeutic toxicity. However, few satisfactory data address the relationships between patterns of lithium toxicity, serum concentrations, and poisoning severity. Therefore, we have prospectively collected data from enquiries to the Scottish Poisons Information Bureau regarding patients with lithium toxicity, and sought to better characterize the clinical features and poisoning severity within each group, and to determine whether lithium concentrations might predict poisoning severity for different patterns of toxicity.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Data from all enquiries to the Scottish Poisons Information Bureau related to lithium toxicity were collected between 1 January 2000 and 31 December 2005. These included age and gender of the patient concerned, the suspected quantity of lithium ingested, co-ingested drugs or alcohol, symptoms and signs, serum lithium concentration, and interval between ingestion and lithium level. Enquiries for information or educational purposes only were excluded. Enquiries from the same or different source related to the same poisoning episode were considered as a single enquiry.
Poisoning severity was determined from clinical data collected at the time of patient referral, using two discrete methods. The Poisoning Severity Score (PSS) was developed jointly by the European Association of Poisons Centres and Clinical Toxicologists, and the International Programme on Chemical Safety and the European Commission. It takes into account the most severe clinical features, and grades severity into: 0, no features; 1, minor; 2, moderate; 3, severe; and 4, fatal poisoning. There is good concordance of PSS scores between centres.24 A specific system for grading the severity of lithium toxicity by clinical features was introduced by Amdisen: 0, no features; 1, mild (nausea, vomiting, tremor, hyperreflexia, agitation, muscle weakness, ataxia); 2, moderate (stupor, rigidity, hypertonia, hypotension); and 3, severe poisoning (myoclonus, collapse, seizures, coma).23
Statistical analyses
Data were considered for each of the three broad patterns of lithium exposure: acute, acute-on-therapeutic, and chronic. Data were not expected to be normally distributed and, therefore, medians and IQRs were used as descriptive statistics. Comparison between groups was made by Mann-Whitney tests for absolute data and Yates’ corrected 
2 tests for proportions, using MedCalc statistical software version 9.0.1.1
[EC]
(www.medcalc.be). Kruskal-Wallis tests were used to examine relationships between lithium concentrations and Poisoning Severity Scores. p values <0.05 were accepted as statistically significant in all cases.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
There were 215 enquiries related to lithium toxicity; 43 were excluded because they were for information only, or related to an earlier enquiry. Therefore, 172 separate patient-related episodes were included in the formal data analyses. Median (IQR) age of patients was 44 (35–54) years, and enquiries related to women in 67.3%. The reported patterns of ingestion were acute in 101 (58.7%), acute-on-therapeutic in 38 (22.1%), and chronic in 33 (19.2%). Median (IQR) stated quantity of lithium ingested was 3750 (1450–9550) mg. Sixty-five patients (37.8%) had co-ingested other drugs or alcohol; these were other antidepressants (52.5%), antipsychotics (34.4%), benzodiazepines (24.6%), ethanol (23.0%), paracetamol (16.4%), and other drugs (49.2%). In 88 patients who ingested lithium alone, patterns of toxicity were acute in 39 (44.3%), acute-on-therapeutic in 23 (26.1%), and chronic in 26 (29.5%).
In patients with chronic toxicity, the stated quantities of lithium ingested were substantially lower, and the interval between ingestion and lithium measurement was longer than in patients with acute and acute-on-therapeutic toxicity (Table 1). Reported serum lithium concentrations were similar between all patterns of lithium toxicity, in the whole population and in the subset that had not co-ingested other drugs or alcohol (Table 2). No significant relationships were found between serum lithium concentration and Poisoning Severity Score in the whole study group, nor in any of the subgroups defined by pattern of lithium toxicity. Clinical features of toxicity were reported more frequently in patients with chronic toxicity, including reduced conscious level and hyperreflexia (Table 3). Moreover, the proportions of patients with moderate and severe poisoning severity scores were higher in chronic and acute-on-therapeutic toxicity compared to acute toxicity (Figure 1).
|
|
|
|
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Patients with chronic lithium toxicity are more likely to exhibit clinical features of toxicity and have higher Poisoning Severity Scores than those with acute toxicity, and acute-on-therapeutic toxicity appears to confer an intermediate risk. This contrasts with the substantially lower quantities of lithium ingested by patients with chronic toxicity (compared to both acute and acute-on-therapeutic toxicity). Serum lithium concentrations were similar across each of the patterns of lithium toxicity, and did not predict poisoning severity or clinical features. Pattern of lithium exposure thus appears to be a stronger indicator of severe toxicity than either the stated quantity of lithium ingested or serum lithium concentrations alone.
Similar patterns have been observed in Emergency Departments. For example, despite acute lithium toxicity accounting for around 76% of hospital admissions with lithium poisoning, moderate to severe poisoning features are present in only 9%, compared to 92% of patients with chronic toxicity.20 The Ontario Poisons Control Center reported a different pattern of referrals due to lithium toxicity, including acute poisoning in 12 patients, acute-on-therapeutic poisoning in 174 patients, and chronic poisoning in 19 patients.25 Despite differences in the case-mix, chronic poisoning was still associated with the greatest risk of toxicity and all were symptomatic, including 17 who had neurological features. Elsewhere, the clinical course after acute lithium toxicity has been described as generally uncomplicated compared to that of patients with chronic toxicity.26,27
One explanation for increased toxicity is that chronic lithium poisoning is likely to be associated with higher tissue concentrations, which might not be readily discernable from serum lithium measurements alone. Patients with acute-on-therapeutic or chronic toxicity may be exposed to high lithium concentrations for longer periods than patients with acute toxicity, due to tissue saturation and impaired lithium clearance. Serum lithium concentrations were evaluated at single time-points, and the present study design does not discriminate differences in the extent of overall lithium exposure between groups. Others have reported a poor correlation or no correlation between serum lithium concentrations and severity of symptoms. More important predictors of toxicity in chronic lithium poisoning include diabetes insipidus, age >50 years, hypothyroidism, and impaired renal function.28 These factors emphasize the importance of duration and extent of lithium exposure in predicting clinically relevant toxicity.
Age is a potential confounding factor, because this differed between patterns of lithium toxicity. Although we did not find correlations between patient age and PSS or Amdisen scores, we cannot exclude the possibility that this may have contributed to poisoning severity. A limitation is that poisoning severity was determined from data collected at the time of patient referral, and might not indicate peak toxicity for individual patients. A further weakness of the data is that selection bias may have been introduced because only patients referred to the Scottish Poisons Information Bureau were evaluated, and our findings might not be generalizable to all lithium poisoning cases. Patients with more severe poisoning or atypical clinical features may have been more likely to have been referred, which might explain a lack of some characteristic adverse effects associated with chronic therapy, for example cerebellar ataxia, slurred speech, and diabetes insipidus. Nonetheless, the data were consistent with patterns and severity of lithium poisoning reported amongst patients admitted to hospital.20 The data collection methods adopted by the Scottish Poisons Information Bureau do not link enquiries to individual patients, and a lack of follow-up outcome data is a limitation of these findings.
In conclusion, our data underpin the greater risk of toxicity conferred by chronic lithium poisoning compared to other patterns of exposure. Neither the stated quantity of lithium ingested nor the serum lithium concentration predicts the likelihood of severe clinical features of poisoning. Clinical management and treatment decisions in lithium poisoning should therefore depend on the pattern of lithium exposure and presence of clinical features of toxicity, rather than laboratory investigations alone.
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. West AP and Meltzer HY. (1979) Paradoxical lithium neurotoxicity: a report of five cases and a hypothesis about risk for neurotoxicity. Am J Psychiatry 136 963–6.
2. Goddard J, Bloom SR, Frackowiak RS, Pusey CD, MacDermot J, Liddle PF. (1991) Lithium intoxication. Br Med J 302 1267–9.[ISI][Medline]
3. Thiel A, Nau R, Lehmann K, Willers T. (1993) Intoxication in manic patients following chaotic self-administration of lithium. Acta Psychiatr Scand 88 289–91.[ISI][Medline]
4. Chang YC, Yip PK, Chiu YN, Lin HN. (1988) Severe generalized polyneuropathy in lithium intoxication. Eur Neurol 28 39–41.[CrossRef][ISI][Medline]
5. Mateer JR and Clark MR. (1982) Lithium toxicity with rarely reported ECG manifestations. Ann Emerg Med 11 208–11.[CrossRef][ISI][Medline]
6. Mitchell JE and Mackenzie TB. (1982) Cardiac effects of lithium therapy in man: a review. J Clin Psychiatry 43 47–51.[ISI][Medline]
7. Fenves AZ, Emmett M, White MG. (1984) Lithium intoxication associated with acute renal failure. South Med J 77 1472–4.[ISI][Medline]
8. Clendeninn NJ, Pond SM, Kaysen G, Barraza JJ, Farrell T, Becker CE. (1982) Potential pitfalls in the evaluation of the usefulness of hemodialysis for the removal of lithium. J Toxicol Clin Toxicol 19 341–52.[ISI][Medline]
9. Schneider JA and Mirra SS. (1994) Neuropathologic correlates of persistent neurologic deficit in lithium intoxication. Ann Neurol 36 928–31.[CrossRef][ISI][Medline]
10. Naramoto A, Koizumi N, Itoh N, Shigematsu H. (1993) An autopsy case of cerebellar degeneration following lithium intoxication with neuroleptic malignant syndrome. Acta Pathol Jpn 43 55–8.[Medline]
11. Peiffer J. (1981) Clinical and neuropathological aspects of long-term damage to the central nervous system after lithium medication. Arch Psychiatr Nervenkr 231 41–60.[CrossRef][ISI][Medline]
12. Mangano WE, Montine TJ, Hulette CM. (1997) Pathologic assessment of cerebellar atrophy following acute lithium intoxication. Clin Neuropathol 16 30–3.[Medline]
13. Bartha L, Marksteiner J, Bauer G, Benke T. (2002) Persistent cognitive deficits associated with lithium intoxication: a neuropsychological case description. Cortex 38 743–52.[ISI][Medline]
14. Singh SV. (1982) Lithium carbonate-fluphenazine decanoate producing irreversible brain damage. Lancet 2 278.[ISI][Medline]
15. West AP. (1985) Lithium-induced neurotoxicity and irreversible brain damage: a possible mechanism and antidote. J Clin Psychiatry 46 300.[ISI][Medline]
16. Apte SN and Langston JW. (1983) Permanent neurological deficits due to lithium toxicity. Ann Neurol 13 453–5.[CrossRef][ISI][Medline]
17. Nagaraja D, Taly AB, Sahu RN, Channabasavanna SM, Narayanan HS. (1987) Permanent neurological sequelae due to lithium toxicity. Clin Neurol Neurosurg 89 31–4.[ISI][Medline]
18. Adityanjee, Munshi KR and Thampy A. (2005) The syndrome of irreversible lithium-effectuated neurotoxicity. Clin Neuropharmacol 28 38–49.[CrossRef][ISI][Medline]
19. Bailey B and McGuigan M. (2000) Comparison of patients hemodialyzed for lithium poisoning and those for whom dialysis was recommended by PCC but not done: what lesson can we learn? Clin Nephrol 54 388–92.[ISI][Medline]
20. Dyson EH, Simpson D, Prescott LF, Proudfoot AT. (1987) Self-poisoning and therapeutic intoxication with lithium. Hum Toxicol 6 325–9.[ISI][Medline]
21. Ferron G, Debray M, Buneaux F, Baud FJ, Scherrmann JM. (1995) Pharmacokinetics of lithium in plasma and red blood cells in acute and chronic intoxicated patients. Int J Clin Pharmacol Ther 33 351–5.[ISI][Medline]
22. Oakley PW, Whyte IM, Carter GL. (2001) Lithium toxicity: an iatrogenic problem in susceptible individuals. Aust N Z J Psychiatry 35 833–40.[CrossRef][ISI][Medline]
23. Hansen HE and Amdisen A. (1978) Lithium intoxication. Report of 23 cases and review of 100 cases from the literature. Q J Med 47 123–44.[ISI][Medline]
24. Persson HE, Sjoberg GK, Haines JA, Pronczuk de Garbino J. (1998) Poisoning severity score. Grading of acute poisoning. J Toxicol Clin Toxicol 36 205–13.[ISI][Medline]
25. Bailey B and McGuigan M. (2000) Lithium poisoning from a poison control center perspective. Ther Drug Monit 22 650–5.[CrossRef][ISI][Medline]
26. Horowitz LC and Fisher GU. (1969) Acute lithium toxicity. N Engl J Med 281 1369.[Medline]
27. Kondziela JR. (1984) Extreme lithium intoxication without severe symptoms. Hosp Community Psychiatry 35 727–8.
28. Oakley PW, Whyte IM, Carter GL. (2001) Lithium toxicity: an iatrogenic problem in susceptible individuals. Aust NZ J Psychiatry 35 833–40.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||