Abstract
Background
The licensed intravenous acetylcysteine regimen for treating paracetamol overdose in most countries uses three separate infusions over 21 h. This complex regimen, requiring different infusion concentrations and rates, has been associated with administration errors. The aim of the present study was to assess the extent of administration delays occurring during this acetylcysteine regimen.
Method
A 6‐month retrospective observational study was conducted at three English teaching hospitals with clinical toxicology services from October 2014. Patients aged 16 years and over, treated with intravenous acetylcysteine for paracetamol overdose, were included. The start times for infusions were recorded and the delays compared with the prescribed infusion times were calculated. Anaphylactoid reactions, intravenous cannula problems, overdose intent and smoking status were recorded to assess their contribution to delays.
Results
From 263 cases identified, 198 met the study inclusion criteria. The median time between the start of infusions 1 and 3 was delayed from the intended 5 h by a median (interquartile range) of 90 (50–163) min, with 135 (68%) cases delayed by more than 1 h. Significantly longer delays were observed in patients with anaphylactoid reactions [median delay 267 (217–413) min, n = 8] and accidental/supratherapeutic overdose [median delay 170 (95–260) min, n = 29]. There were no significant differences between smokers and nonsmokers, or for patients with intravenous cannula problems.
Conclusion
Long delays were identified during the three‐infusion acetylcysteine regimen for the treatment of paracetamol overdose. These were of clinical significance and could lead to periods of subtherapeutic plasma acetylcysteine concentrations and potentially avoidable hepatotoxicity, as well as delaying hospital discharge.
Keywords: acetaminophen overdose, acetylcysteine, drug administration, intravenous infusions, paracetamol overdose
What is Already Known about this Subject
The licensed intravenous acetylcysteine regimen for treating paracetamol overdose in most countries uses three separate infusions, administered over 21 h.
This complex regimen, requiring different infusion concentrations and rates, has been associated with administration errors.
What this Study Adds
Long delays were identified during the three‐infusion acetylcysteine regimen for the treatment of paracetamol overdose.
Delays could increase the length of hospital stay and were of clinical significance, and could lead to subtherapeutic plasma acetylcysteine concentrations and potentially avoidable hepatotoxicity.
Table of Links
This Table lists key ligands in this article which are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY 1 and are permanently archived in the Concise Guide to PHARMACOLOGY 2015/16.
Introduction
In most countries, the licensed intravenous acetylcysteine regimen for treating paracetamol (acetaminophen) overdose involves three separate infusions, given over a total of 21 h. The first infusion is administered over 1 h, the second over 4 h and the third (and any subsequent) infusions over 16 h 2. Each infusion is prepared by adding the required volume of acetylcysteine solution to three different volume infusion bags of 5% dextrose or 0.9% sodium chloride. There should ideally be no delay in between the infusions, to minimize treatment duration and length of stay, while providing maximal hepatic protection from the toxic metabolite of paracetamol, N‐acetyl‐p‐benzoquinone imine (NAPQI). Treatment is usually started in the emergency department and continued on an admission/medical/observation ward.
The complexity of the regimen, with the requirement for three different infusion concentrations and rates, has been associated with administration errors 3, 4, 5. In a US retrospective chart review, Hayes and colleagues found medication errors in the treatment of 33% of 221 patients treated with intravenous acetylcysteine for paracetamol overdose 4. The most frequent error, in 18.6% cases, was the interruption of treatment by more than 1 h; in addition, incorrect infusion rates were seen in 5% of cases (the authors did not record whether administration was too fast or slow) 4. In a Malaysian audit of 236 patients treated with acetylcysteine for paracetamol poisoning, interruptions in treatment of more than 1 h were seen in fewer cases (5.5%) than in the US study but infusion rate errors were recorded in more (37.3%) 5. In the US study, the same three‐infusion regimen was in place as currently used in the UK. In the Malaysian study, the three‐infusion regimen was the same, apart from the initial infusion being administered over 15 min, rather than 1 h, as was routine in the UK until 2012.
Consistent with these US and Malaysian studies, our own experience is that the time taken for acetylcysteine infusions to be completed is often longer than expected. The aim of the present study was to assess the extent of delays during the administration of the three‐infusion intravenous acetylcysteine regimen in three English hospitals, and to understand when these delays occurred during treatment.
Methods
A retrospective observational study was conducted at three English teaching hospitals with clinical toxicology services. Patients aged 16 years and over, treated with intravenous acetylcysteine for paracetamol overdose between 1 October 2014 and 31 March 2015, were identified from the clinical toxicology databases used at the three hospitals. Patients were excluded if any of the start times for the three infusions were not recorded or if treatment was stopped prior to the start of the third infusion.
In addition to basic demographic data (age/gender), the start/stop times for all infusions were recorded and delays to treatment (in min) were calculated, comparing times recorded for administration with the predicted times based on the prescribed infusion times. This was recorded for: (i) the start of the first and second infusions; (ii) the start of the second and third infusions; and (iii) the start of the first and third infusions. The primary outcome measure was the delay between the start of the first and the start of the third infusion. When documented, infusion stop times were also recorded and the total delays to complete all three infusions were calculated. In addition, smoking status, overdose intent (intentional or accidental/supratherapeutic overdose), anaphylactoid reactions (defined as documented anaphylactoid reaction, bronchospasm, rash, swelling or reaction requiring treatment with antihistamines and/or steroids) and intravenous cannula‐related problems (documentation that an intravenous cannula was misplaced or stopped working) were recorded to assess if these contributed to any delays that occurred.
The data were collated in an Excel spreadsheet and analysed using Excel and SPSS Statistics (version 21, IBM, Armonk, NY, USA). The impact of the additional factors was assessed using Mann–Whitney U tests. Results from the three hospitals were compared using Kruskal–Wallis tests.
Data for auditing the management of paracetamol overdose are routinely and prospectively collected on databases held within the clinical toxicology units at the study centres, and these are approved by the local data protection officers/Caldicott guardians.
Results
There were 263 cases identified during the 6‐month study period at the three hospitals: 86 in London (15 excluded), 108 in Newcastle (31 excluded) and 69 in York (19 excluded). Sixty‐five cases were excluded because treatment was discontinued before the start of the third infusion (19 cases), the start times for all three infusions were not documented (25 cases) or the prescription charts or notes were missing (21 cases). Thus, 198 cases were included in the analysis, with a mean ± standard deviation (SD) age of 33.6 ± 15.9 years and 114 (57.5%) were female. Those excluded had a mean ± SD age of 32.8 ± 13 years and 54.0% were female.
The median [interquartile range (IQR)] delay compared with the 5 h interval intended between the start of infusions 1 and 3 was 90 (50–163) min. The median delay compared with the 1 h interval intended between the start of infusions 1 and 2 was 25 (10–60) min and compared with the 4 h interval intended between the start of infusion 2 and 3 was 50 (15–104) min. There was a delay of more than 1 h compared with the prescribed times between the start of infusions 1 and 3 in 135 (68.2%) cases, of more than 2 h in 78 (39.4%) cases and of more than 3 h in 41 (20.7%) cases. In four cases (2.0%), there were delays of more than 10 h compared with prescribed times between the start of infusions 1 and 3.
Smoking status, anaphylactoid reactions and cannula problems were recorded at the hospitals in London and Newcastle but not York. Comparisons between smokers and nonsmokers, intentional and accidental/supratherapeutic overdoses, and those with/without cannula problems or anaphylactoid reactions are shown in Table 1. Smoking status had no significant effect on the delays but patients with accidental/supratherapeutic overdoses or anaphylactoid reactions experienced significantly longer delays.
Table 1.
Comparisons between different groups for median delays between the start of infusions 1 to 3
| Median (IQR) delay between start of infusion1 and start of infusion 2 (min) | Median (IQR) delay between start of infusion 2 and start of infusion 3 (min) | Median (IQR) delay between start of infusion 1 and start of infusion 3 (min) | P‐value | |
|---|---|---|---|---|
| Intentional overdose (n = 169) | 20 (6–50) | 45 (15–93) | 85 (45–145) | <0.001 |
| Accidental/supratherapeutic overdose (n = 29) | 80 (35–135) | 75 (30–120) | 170 (95–260) | |
| Smoker (n = 72) | 20 (10–55) | 50 (25–113) | 90 (50–185) | NS |
| Nonsmoker (n = 59) | 30 (5–72) | 60 (8–110) | 90 (60–168) | |
| Anaphylactoid reaction (n = 8) | 188 (140–230) | 60 (26–93) | 267 (217–413) | <0.001 |
| No anaphylactoid reaction (n = 147) | 25 (10–60) | 55 (15–108) | 90 (53–155) | |
| Cannula problem (n = 11) | 20 (10–80) | 50 (43–103) | 150 (35–198) | NS |
| No cannula problem (n = 136) | 25 (10–62) | 55 (15–107) | 90 (54–164) |
Groups were compared using the Mann–Whitney U test. Note: smoking status, anaphylactoid reactions and cannula problems were not recorded for all patients
IQR, interquartile range; NS, not significant
Adverse effects were recorded in London and Newcastle. Eight (5.4%) patients had anaphylactoid reactions and 28 (18.9%) experienced nausea and/or vomiting. The median (IQR) delay between the start of infusions 1 and 3 compared with prescribed times for patients with anaphylactoid reactions was 267 (217–413) min compared with 90 (53–155) min for patients without anaphylactoid reactions (n = 147). Delays for patients with anaphylactoid reactions were 188 (140–230) min between the start of infusions 1 and 2 and 60 (26–93) min between the start of infusions 2 and 3. At the two centres where adverse reactions were recorded, 34 (23.0%) patients experienced delays of more than 3 h and six (17.6%) of these had anaphylactoid reactions. For patients without anaphylactoid reactions, the median (IQR) delay from the start of infusion 1 to the start of infusion 3 remained significantly longer for those with accidental/supratherapeutic overdose; 155 (90–255) min compared with 85 (50–142) min for intentional overdoses (P = 0.0008).
The stop time for the third infusion was recorded in only 36 (18.2%) cases. For these patients, the median (IQR) delay between the start of infusion 1 and the end of infusion 3 was 175 (103–300) min. The median (IQR) delay compared with the 16 h interval intended from the start of infusion 3 to the end of infusion 3 for this group was 60 (20–139) min. The median (IQR) delay between the start of infusions 1 and 3 for this group was 75 (45–152) min, which was not significantly different to the 90 min delay observed in the whole cohort.
The delays compared with prescribed times in each of the three hospitals are demonstrated in Figure 1. Comparing the three hospitals, there was a median (IQR) delay between the start of infusions 1 and 3 of 140 (67–197) min in London, 75 (40–130) min in Newcastle and 90 (45–145) min in York. Delays in Newcastle were significantly shorter than in London (P = 0.0015). Differences between delays comparing Newcastle and York, and London and York were not significant.
Figure 1.
The length of delays compared with prescribed times from the start of infusion 1 to the start of infusion 3 for the three hospitals
Discussion
A delay of more than 1 h compared with prescribed infusion times between the start of infusions 1 and 3 occurred for over two‐thirds of patients in the present study. For the group with total infusion times recorded, further delays occurred both between each infusion and during the infusions. This suggests that the delays result from both the time preparing and instigating infusions, and during infusions.
The impact of delays in administration on plasma concentrations of acetylcysteine and outcomes has not previously been reported, but Hayes and colleagues 4 suggested that delays of more than 1 h should be considered potentially significant based on the elimination half‐life for acetylcysteine of 5.7 h found by Prescott et al. 6 in a pharmacokinetic study of the three‐infusion regimen. The extent of the delays identified could lead to potentially avoidable hepatotoxicity. However, it is not clear at what point the duration of the delay might lead to an increased risk of liver injury.
Infusion delays prolong inpatients stays, increase costs and increase acute medical inpatient bed occupancy. Most patients (85% of this cohort) are treated following intentional paracetamol overdoses, and prolonging their stay as a medical inpatient delays psychiatric/psychosocial assessment.
Anaphylactoid reactions and intravenous cannula‐related problems contributed to delays in treatment but there were important delays for patients with no documented evidence of either of these complications. The incidence of anaphylactoid reactions in the present study was 5.2%, which is lower than previously reported 7, 8, 9, 10. The incidence of intravenous cannula‐related problems was also lower than expected; these included removal, difficulty in reinserting and blockages in tubing. The retrospective data collection and reliance on documentation of anaphylactoid reactions and cannula problems in the medical notes may have resulted in under‐reporting of these events.
It was expected that patients leaving the ward to smoke might delay infusions more in smokers compared with nonsmokers. However, smoking status had no effect on the delays. This may have been because of the use of nicotine replacement in hospital inpatients.
Surprisingly, delays were significantly longer for patients presenting following accidental overdose compared with those with intentional overdoses. Patients with lower serum paracetamol concentrations are at greater risk of anaphylactoid reactions 10. Those with accidental/supratherapeutic overdose often present with lower serum paracetamol concentrations and therefore a potentially higher rate of adverse reactions in this group could have contributed to their delays. However, delays remained longer in those who did not suffer anaphylactoid reactions, suggesting that other factors were important. It is possible that less severe reactions occurred for some patients and delayed infusions, but this was not documented. Patients presenting following intentional overdoses frequently require close observation by staff, and occasionally 1:1 observation by a mental health nurse. It is possible that the need for increased observation results in earlier recognition of acetylcysteine infusions finishing, or problems with the infusions. Delays could potentially have been caused by waiting for repeat blood test results in deciding on the treatment courses for patients with staggered overdoses. However, local practice is to continue infusions until results are available (up until the end of the third infusion).
Patients are usually transferred between departments at least once during the course of their treatment with acetylcysteine (most often between the emergency department and admission/medical/observation wards). The prolonged nature of the infusion involves the handover of care between medical and nursing staff working in shifts. There may have been delays in starting the next infusion while care was handed over between teams.
Electronic infusion pumps are used to set infusion times. Therefore, theoretically, administration during infusions should not be delayed. However, infusion rate errors have been noted in previous US and Malaysian studies on acetylcysteine errors, although the exact nature of these errors was not clear 4, 5. Infusion pump calibrations were not assessed as part of the current study. Discrepancies in the volumes of infusions could also have contributed to delays. From personal communication with the manufacturer for the infusion bags used in London, the volumes in the infusion bags had the following ranges: 265–277 ml (250 ml bag), 520–540 ml (500 ml bag) and 1025–1069 ml (1000 ml bag). Infusion times would have been longer than expected if the infusion pumps had not been set to account for these volumes and the volume of acetylcysteine added. However, a small volume of infusion fluid remains in the tubing and infusion bag on completion of the infusion. These factors could contribute to short delays but would not have resulted in the magnitude of delays found in the present study.
With stop times inadequately recorded in the present study, it is difficult to ascertain whether the delays occurred mostly between infusions or during infusions. Our study demonstrated longer delays between the start of infusions 2 and 3 compared with between the start of infusions 1 and 2 (in view of the longer duration of infusion 2); this suggests that delays are likely also to have occurred during the infusions rather than simply between ending one infusion and starting the next. Delays for patients with anaphylactoid reactions were longer between the start of infusions 1 and 2 compared with between the start of infusions 2 and 3. This suggests that reactions may have been more frequent or severe during the first infusion compared with the second. However, there were few patients with documented anaphylactoid reactions, limiting the interpretation of this finding.
Acetylcysteine dosing tables produced by the Medicines and Healthcare products Regulatory Agency, as part of the 2012 changes to UK guidelines on the treatment of paracetamol overdose, were used in all three centres 2. These were produced with the aim of simplifying drug calculations and preparation, and providing the volume of acetylcysteine to be added for each infusion.
Delays in the initiation of treatment also affects the timely administration of acetylcysteine for paracetamol overdose. This was not assessed in the present study. The Royal College of Emergency Medicine Paracetamol Overdose Clinical Audit 2013–2014 found that, for patients presenting less than 8 h from ingestion, 50% received treatment with acetylcysteine within the recommended 8 h 11. For patients presenting more than 8 h after ingestion, 80% of emergency departments did not administer acetylcysteine to any patients within the recommended 1 h 11. In a single centre audit, Pettie and colleagues found that, for patients presenting either 8–24 h postingestion or with staggered overdose and considered at risk of hepatotoxicity, 12% had acetylcysteine started within 90 min of arrival, and this improved to 61% following the introduction of an integrated care pathway 12. These authors also found that the introduction of the integrated care pathway had led to improvements in blood sampling and treatment decisions, and to a reduction in prescription errors.
To improve administration times, these issues should be highlighted to those treating patients with intravenous acetylcysteine. During nursing and medical handovers, the prescribed start and finish times for each infusion could be reviewed and infusion times monitored. All three centres in the current study currently use paper prescription charts of acetylcysteine; electronic prescribing systems could aid this process, including medication administration timing reminders or alerts to notify staff that there has been a delay in the infusion time. Electronic prescribing should also improve the documentation of infusion times, enabling times to be standardized/coordinated rather than relying on individual watches/clocks.
Preparation of the second and third infusions immediately after the first has started, so they can be changed more quickly could reduce the delays between infusions. Improved documentation of infusion stop times would help to identify which steps are associated with the longest delays and determine the extent of the delays for the full three infusions. If a delay is recognized, the infusion rate could be increased to target the planned infusion stop time. This is unlikely to increase the number of adverse effects because these occurred at similar rates when the first infusion was previously administered over 15 min in the UK 13.
Recently, a number of alternative two‐infusion acetylcysteine regimens have been trialled, and these are associated with a lower incidence of adverse effects 14, 15, 16, 17, 18. Reducing the number of infusions required is also likely to reduce infusion delays. The Scottish and Newcastle Antiemetic Pre‐treatment (SNAP) regimen used by Bateman and colleagues 14 is shorter (with a duration of 12 h) and therefore would significantly reduce total infusion times, irrespective of the delays. However, despite promising initial results, further evidence is required to demonstrate the efficacy of this shorter regimen.
Limitations
The present study relied on retrospective data collection for identifying infusion start times; it is possible these may not have reflected the actual start time of the infusions. Data collection was from handwritten medical notes, with times recorded from nonstandardized clocks. As a result of inadequate recording of stop times, the delays reported in most cases were between the start of the first and third infusions, and therefore would have under‐represented the total extent of delays. Markers of hepatotoxicity, paracetamol concentrations and patient outcomes were not recorded in the present study.
Conclusion
Long delays were identified during the three‐infusion acetylcysteine regimen for the treatment of paracetamol overdose. Delays increase the length of hospital stay and were found to be of clinical significance, potentially leading to subtherapeutic plasma acetylcysteine concentrations and avoidable hepatotoxicity. Early preparation of infusions, adjusting infusion times to compensate for delays, and novel regimens using two infusions are options that may reduce delays.
Competing Interests
All authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: SHLT and PID are members of the Commission on Human Medicines Paracetamol Expert Group. There are no other competing interests.
Contributors
P.D., D.W. and J.A. conceived the study. G.B., J.N., M.E. and W.S.W. collected the data. G.B. collated and analysed the data. All authors contributed to reviews and approved the final manuscript.
Bailey, G. P. , Najafi, J. , Elamin, M. E. M. O. , Waring, W. S. , Thomas, S. H. L. , Archer, J. R. H. , Wood, D. M. , and Dargan, P. I. (2016) Delays during the administration of acetylcysteine for the treatment of paracetamol overdose. Br J Clin Pharmacol, 82: 1358–1363. doi: 10.1111/bcp.13063.
References
- 1. Southan C, Sharman JL, Benson HE, Faccenda E, Pawson AJ, Alexander SP, et al. The IUPHAR/BPS Guide to PHARMACOLOGY in 2016: towards curated quantitative interactions between 1300 protein targets and 6000 ligands. Nucleic Acids Res 2016; 44 (Database Issue): D1054–D1068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Commission on Human Medicines . Paracetamol overdose: new guidance on the use of intravenous acetylcysteine [online]. Available at http://webarchive.nationalarchives.gov.uk/20141205150130/http://www.mhra.gov.uk/home/groups/pl‐p/documents/drugsafetymessage/con178654.pdf (last accessed 10 February 2016).
- 3. Ferner RE, Langford NJ, Anton C, Hutchings A, Bateman DN, Routledge PA. Random and systematic medication errors in routine clinical practice: a multicentre study of infusions, using acetylcysteine as an example. Br J Clin Pharmacol 2001; 52: 573–577. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Hayes BD, Klein‐Schwartz W, Doyon S. Frequency of medication errors with intravenous acetylcysteine for acetaminophen overdose. Ann Pharmacother 2008; 42: 766–770. [DOI] [PubMed] [Google Scholar]
- 5. Au V, Zakaria MI. A study on the medication errors in the administration of N‐acetylcysteine for paracetamol overdose patients in Malaysia. Hong Kong J Emerg Med 2014; 21: 362–367. [Google Scholar]
- 6. Prescott LF, Donovan JW, Jarvie DR, Proudfoot AT. The disposition and kinetics of intravenous N‐acetylcysteine in patients with paracetamol overdosage. Eur J Clin Pharmacol 1989; 37: 501–506. [DOI] [PubMed] [Google Scholar]
- 7. Yamamoto T, Spencer T, Dargan PI, Wood DM. Incidence and management of N‐acetylcysteine‐related anaphylactoid reactions during the management of acute paracetamol overdose. Eur J Emerg Med 2014; 21: 57–60. [DOI] [PubMed] [Google Scholar]
- 8. Schmidt LE. Identification of patients at risk of anaphylactoid reactions to N‐acetylcysteine in the treatment of paracetamol overdose. Clin Toxicol 2013; 51: 467–472. [DOI] [PubMed] [Google Scholar]
- 9. Lynch RM, Robertson R. Anaphylactoid reactions to intravenous N‐acetylcysteine: a prospective case controlled study. Accid Emerg Nurs 2004; 12: 10–15. [DOI] [PubMed] [Google Scholar]
- 10. Pakravan N, Waring WS, Sharma S, Ludlam C, Megson I, Bateman DN. Risk factors and mechanisms of anaphylactoid reactions to acetylcysteine in acetaminophen overdose. Clin Toxicol 2008; 46: 697–702. [DOI] [PubMed] [Google Scholar]
- 11. Royal College of Emergency Medicine . Paracetamol Overdose Clinical Audit 2013–2014 [online]. Available at https://secure.rcem.ac.uk/CEM/document?id=8120 (last accessed 20 June 2016).
- 12. Pettie JM, Dow MA, Sandilands EA, Thanacoody HKR, Bateman DN. An integrated care pathway improves the management of paracetamol poisoning. Emerg Med J 2012; 29: 482–486. [DOI] [PubMed] [Google Scholar]
- 13. Bateman DN, Carroll R, Pettie J, Yamamoto T, Elamin ME, Peart L, et al. Effect of the UK's revised paracetamol poisoning management guidelines on admissions, adverse reactions and costs of treatment. Br J Clin Pharmacol 2014; 78: 610–618. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Bateman DN, Dear JW, Thanacoody HK, Thomas SH, Eddleston M, Sandilands EA, et al. Reduction of adverse effects from intravenous acetylcysteine treatment for paracetamol poisoning: a randomised controlled trial. Lancet 2014; 383: 697–704. [DOI] [PubMed] [Google Scholar]
- 15. Wong A, Graudins A. A Simplification of the standard three‐bag intravenous acetylcysteine regimen for paracetamol poisoning results in a lower incidence of adverse drug reactions. Clin Toxicol 2016; 54: 115–119. [DOI] [PubMed] [Google Scholar]
- 16. Isbister GK, Downes MA, McNamara K, Berling I, Whyte IM, Page CB. A prospective observational study of a novel 2‐phase infusion protocol for the administration of acetylcysteine in paracetamol poisoning. Clin Toxicol 2016; 54: 120–126. [DOI] [PubMed] [Google Scholar]
- 17. Bateman DN, Dear JW, Thomas SH. New regimens for intravenous acetylcysteine, where are we now? Clin Toxicol 2016; 54: 75–78. [DOI] [PubMed] [Google Scholar]
- 18. Waring WS. Novel acetylcysteine regimens for treatment of paracetamol overdose. Ther Adv Drug Saf 2012; 3: 305–315. [DOI] [PMC free article] [PubMed] [Google Scholar]