Abstract
INTRODUCTION:
Acetaminophen (APAP) toxicity is the main cause of acute liver failure in the United States. A prior series (1992–1995) identified 71 hospitalized adults with APAP toxicity through the International Statistical Classification of Disease and Related Health Problems, 9th revision (ICD-9) code at Parkland Hospital, Dallas, TX.
METHODS:
We used a laboratory database search of serum APAP levels from 2011 to 2015 to identify patients with APAP toxicity in the same hospital.
RESULTS:
We identified 140 patients hospitalized for APAP toxicity from 27,143 APAP levels obtained; 35 required Intensive Care Unit (ICU) admission, and there were no deaths. APAP toxicity/100,000 admissions was similar between eras.
DISCUSSION:
APAP toxicity continues unabated after 20 years but with improved overall outcomes.
INTRODUCTION
Acetaminophen (APAP) toxicity, the most common cause of acute liver failure (ALF), accounts for ~46% of ALF in Europe and North America (1). A 1997 study described the prevalence and characteristics of APAP overdose at Parkland Hospital, Dallas, Texas, an urban county safety-net hospital, using International Statistical Classification of Disease and Related Health Problems, 9th revision (ICD-9) codes (2); 71 patients hospitalized over a 40-month period with actual or potential APAP toxicity were divided into intentional (suicide attempt) or unintentional (overuse of APAP for relief of pain/fever, etc). Generally, unintentional patients presented late, suffered more severe injury, and were more likely to die because of toxicity (4 of 21 unintentional vs 1 of 50 intentional). Ten patients had ALF, and 5 died; 23 had aminotransferases ≥ 1,000 IU/L, and 56 required N-acetylcysteine, a known safe antidote (3-6).
Twenty years later, we used the electronic medical record and laboratory database of serum APAP levels to identify APAP-related admissions. We hypothesized that we would be able to encapture a greater number of APAP toxicity events through this search as compared with a prior International Statistical Classification of Disease and Related Health Problems (ICD) code review, given discrepancies in accurate coding. We reviewed in detail circumstances of the overdose, biochemical features, and clinical outcomes to
1. determine the incidence of APAP toxicity during the study period using APAP testing,
2. evaluate changes in incidence, clinical features, and outcomes of APAP toxicity in the same setting 20 years later,
3. estimate the economic burden of APAP toxicity in a safety-net hospital setting.
METHODS
In this retrospective cohort study, we compared the current results with the 1997 study. Instead of ICD-9 codes, we identified patients by reviewing APAP testing and levels during the 5-year period, using the same analysis criteria as in the earlier study: a history of APAP ingestion plus any detectable APAP level (≥10 μg/mL) or history plus elevated aminotransferase levels ≥ 1,000 IU/L without a detectable APAP level (APAP <10 μg/mL) (Figure 1). Two groups were identified:
Figure 1.
Selection Criteria Schematic
Group 1: APAP detectable and documentation of a large APAP ingestion per patient history.
Group 2: At least moderate acute liver injury (aminotransferase levels >1,000 IU/L) with no detectable APAP level and documentation of large APAP exposure.
We excluded other etiologies, such as ischemic hepatopathy, and those with insufficient laboratory data or historical evidence to confirm APAP ingestion and then limited the detailed analysis to hospitalized patients (Figure 1).
RESULTS
APAP levels were measured 27,143 times over the 5-year period during 25,952 unique patient encounters (Figure 1); 1,191 levels represented multiple testing for the same patient during the same hospital stay (range 2–23 tests/patient). Only 238 APAP tests (0.87% of all tests) were supratherapeutic (upper limit of normal (ULN); ≥ 30 μg/mL) among the 25,952 unique patients tested (0.92%), the highest APAP level recorded being 471.5 μg/mL. An additional 188 cases had milder degrees of injury or none but required emergency department (ED) observation and/or treatment only; 43 of these (23%) received at least 1 dose of NAC before discharge.
Of the 140 patients, 129 patients (92%) were in group 1 and 11 (8%) in group 2 (Figure 1). Among 78 patients with elevated aminotransferases, 27 had peak levels > 1,000 IU/L and 18 had peak levels >3,500 IU/L; 109 patients (77.8%) had attempted suicide (intentional overdose) while 29 (20.7%) were considered to have unintentional APAP ingestions (Figure 1).
Comparison of intentional vs unintentional APAP toxicity
Between the 2 phenotypes of APAP toxicity, intentional overdose patients were significantly younger (median age 32 vs 38 years), had ingested larger amounts of APAP (median 14 vs 7.5 g/d for unintentional), arrived at the ED earlier after the ingestion (median 3 vs 24 hours), had higher peak serum APAP levels (median 58.3 vs 30.2 μg/L), and had lower peak aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels (median 31 vs 186 and 33 vs 243 IU/L, respectively) (Table 1). Unintentional patients required a significantly longer hospital stay (median 4 vs 3 days, P < 0.001) and higher total charged hospital costs (median $37,566 vs $21,789, P = 0.0145). The overall charged cost of these hospitalizations over the 5-year period was estimated at $5,701,378.
Table 1.
Baseline characteristics of study population, laboratory values, and outcomes
| Variable | All patients (N = 140) | Intentional overdosea (N = 109) | Unintentional overdosea (N = 29) | P value | |||
|---|---|---|---|---|---|---|---|
| Patients | Value | Patients | Value | Patients | Value | ||
| Median age (Q1–Q3) | 140 | 35 (24.75–36) | 109 | 32 (24–46) | 29 | 38 (30–50) | 0.048 |
| Sex—no. (%) | 140 | 109 | 29 | 0.53 | |||
| Female | 79 (56.4) | 60 (55.0) | 18 (62.1) | ||||
| Male | 61 (43.5) | 49 (45.0) | 11 (37.9) | ||||
| Race—no. (%) | 140 | 109 | 29 | 0.42 | |||
| White/Hispanic | 110 (78.5) | 97 (79.8) | 22 (74.9) | ||||
| Black | 26 (18.5) | 20 (18.3) | 5 (17.2) | ||||
| Asian | 3 (2.14) | 1 (0.9) | 2 (6.9) | ||||
| American Indian | 0 (0) | 0 (0) | 0 (0) | ||||
| Others | 1 (0.7) | 1 (0.9) | 0 (0) | ||||
| Prior psych history—no. (%) | 136 | 93 (68.4) | 109 | 78 (71.6) | 27 | 14 (51.9) | 0.07 |
| Median Grams of APAP ingested (Q1–Q3) | 82 | 10 (6–20.75) | 64 | 14 (6.38–24.25) | 18 | 7.5 (5.48–8.38) | 0.019 |
| Median time PTA in hr (Q1–Q3) | 80 | 3 (1–8.25) | 71 | 3 (1–6.5) | 9 | 24 (18.375–30) | 0.044 |
| PTA < 4 h—no. (%) | 80 | 44 (55) | 71 | 41 (57.7) | 9 | 3 (33.3) | 0.29 |
| PTA ≥ 24 h—no. (%) | 80 | 9 (11.2) | 71 | 3 (4.2) | 9 | 6 (66.7) | <0.001 |
| Products consumed—no. (%) | 128 | 102 | 26 | 0.004 | |||
| Single product | 64 (50) | 43 (42.1) | 21 (80.7) | ||||
| Multiple products | 64 (50) | 59 (57.8) | 5 (19.2) | ||||
| Other coingestants taken—no. (%) | 128 | 64 (50) | 103 | 59 (57.2) | 26 | 5 (19.2) | <0.001 |
| Concurrent alcohol ingestion—no. (%) | 134 | 36 (27.8) | 107 | 24 (22.4) | 26 | 11 (42.3) | 0.038 |
| Median peak APAP level in μg/L (Q1–Q3) | 140 | 46.65 (23.6–114.45) | 109 | 58.3 (24.7–123.5) | 29 | 30.2 (15.9–73.8) | 0.013 |
| Peak APAP—no. (%) | 140 | 109 | 29 | ||||
| ≥50 μg/L | 68 (48.5) | 61 (52.2) | 10 (34.5) | 0.06 | |||
| ≥100 μg/L | 43 (30.7) | 38 (34.8) | 5 (17.2) | 0.08 | |||
| Median peak AST in IU/L(Q1–Q3) | 140 | 35 (23–186) | 109 | 31 (21–67) | 29 | 186 (53–2,655) | <0.001 |
| Median peak ALT in IU/L (Q1–Q3) | 140 | 37.5 (19–201.75) | 109 | 33 (18–69) | 29 | 243 (70–2,935) | <0.001 |
| Peak AST—no. (%) | 140 | 109 | 29 | ||||
| >1,000 IU/L | 25 (17.8) | 12 (11) | 12 (41.4) | <0.001 | |||
| >3,500 IU/L | 15 (10.7) | 8 (7.3) | 6 (20.7) | 0.08 | |||
| Peak ALT—no. (%) | 140 | 109 | 29 | ||||
| >1,000 IU/L | 26 (18.6) | 13 (11.9) | 12 (41.4) | <0.001 | |||
| >3,500 IU/L | 14 (10.0) | 10 (8.8) | 5 (17.2) | 0.30 | |||
| Peak AST or ALT—no. (%) | 140 | 109 | 29 | ||||
| >1,000 IU/L | 27 (19.0) | 13 (11.9) | 13 (44.8) | <0.001 | |||
| >3,500 IU/L | 28 (13.0) | 10 (9.1) | 7 (24.1) | 0.05 | |||
| Median peak Bilirubin in mg/dL (Q1–Q3) | 139 | 0.5 (0.3–0.8) | 109 | 0.4 (0.3–0.7) | 28 | 0.6 (0.3–2.15) | 0.22 |
| Median peak INR (Q1–Q3) | 121 | 1.2 (1.1–1.3) | 95 | 1.2 (1.1–1.3) | 24 | 1.3 (1.2–1.725) | 0.008 |
| N-acetylcysteine therapy—no. (%) | 140 | 119 (85) | 109 | 93 (78) | 29 | 24 (82.8) | 0.73 |
| Median length of stay in d (Q1–Q3) | 140 | 3 (2–5) | 109 | 3 (2–4) | 29 | 4 (2–7) | 0.009 |
| ICU patients—no. (%) | 140 | 35 (25) | 109 | 24 (22) | 29 | 9 (31.0) | 0.31 |
| Hepatic coma present—no. (%) | 135 | 5 (3.7) | 106 | 2 (1.8) | 28 | 2 (7.1) | 0.19 |
| Median cost in dollars (Q1–Q3) | 140 | 23,132 (16,851–39,027) | 109 | 21,789 (16,295–34,057) | 29 | 37,566 (19,133–73,399) | 0.015 |
| No. of 1997 criteria met—no. (%) | 140 | 109 | 29 | 0.018 | |||
| 1 | 7 (5) | 3 (2.9) | 4 (13.8) | ||||
| 2 | 122 (87) | 99 (90.8) | 21 (72.4) | ||||
| 3 | 11 (0.7) | 7 (6.4) | 4 (13.8) | ||||
Q1–Q3: indicates range from quartile 1 to quartile 3 of the data set.
ALT, alanine aminotransferase; AST, aspartate aminotransferase; APAP, acetaminophen; ICU, intensive care unit; INR, international normalized ratio; No., number; PTA, prior to arrival.
Indicates 2 subjects were of unknown intent.
Comparison with the earlier study (1992–1995)
The number of patients with APAP toxicity hospitalized per 100,000 adult admissions during increased between the 2 eras from 60/100,000 during the 1990s to 73.4/100,000 adult admissions during 2011–2015. The current patient group had shorter lengths of stay (median 3 vs 4 days), hepatic encephalopathy was much less frequent, and there were no deaths compared with 5 deaths among 71 patients (7%) in the 1997 study. The current APAP toxicity cases accrued much higher total median costs (median $23,132 vs $11,119, when adjusted for inflation) despite their overall shorter lengths of stay (Table 2).
Table 2.
Comparison of historical vs present data
| 1992–1995 | 2011–2015 | P value | |||
|---|---|---|---|---|---|
| N | Values | N | Values | ||
| Median age (Q1–Q3) | 71 | 28 (21.75–37.5) | 140 | 35 (24.75–46) | <0.001 |
| Sex (female/male) | 71 | 48/23 | 140 | 79/61 | 0.16 |
| NAC given—no. (%) | 64 | 56 (87.5) | 140 | 119 (85) | 0.64 |
| Median length of stay in d (Q1–Q3) | 71 | 4 (3–5) | 140 | 3 (2–5) | 0.044 |
| Hepatic coma—no. (%) | 71 | 10 (14) | 135 | 5 (3.5) | 0.012 |
| Death—no. (%) | 71 | 5 (7) | 140 | 0 (0) | 0.007 |
| Median peak APAP level in μg/L (Q1–Q3) | 64 | 104 (42–178.5) | 140 | 46.7 (23.6–114.5) | 0.006 |
| Median peak AST in IU/L (Q1–Q3) | 70 | 33 (19.25–806) | 140 | 35 (23–186) | 0.94 |
| Median peak ALT in IU/L (Q1–Q3) | 61 | 61.5 (16.25–2,515) | 140 | 37.5 (19–201.75) | 0.37 |
| Median peak INR (Q1–Q3)b | 52 | 1.19 (1.07–1.60) | 121 | 1.2 (1.1–1.3) | 0.29 |
| Median cost (in dollars)a | 63 | 11,119 | 140 | 23,132 | <0.001 |
Q1-Q3: Indicates range from quartile 1 to quartile 3 of the data set.
ALT, alanine aminotransferase; AST, aspartate aminotransferase; APAP, acetaminophen; INR, international normalized ratio; No., number.
1992–1995 cost adjusted for inflation by a factor of 68%.
199–1995 data adjusted from peak prothrombin time to INR by a factor of 12.
DISCUSSION
Hepatotoxicity due to APAP use still affects the healthcare system (7,8). Several differences were observed between the 2 eras. The 13% increase in APAP hospitalizations between the 2 eras may reflect the different search strategies resulting in increased case ascertainment. However, the overall number of presumed APAP cases over the second 5-year era was actually 328 rather than the 140 hospitalized patients we analyzed. Currently, many potential overdoses are managed solely in the ED instead of requiring admission. Thus, more than 50% of our documented current overdose cases did not require full hospital admission but still used hospital resources. During both eras, intentional overdoses were more frequent than unintentional ones, whereas unintentional overdose patients experienced more severe injury, including higher aminotransferase and international normalized ratio (INR) values and longer hospital stays.
Importantly, patients requiring admission fared better, with no deaths out of 140 cases versus 5 deaths in 71 cases from the earlier era (P < 0.003). Nonetheless, APAP overdoses continue to result in high resource utilization, beginning with the initial testing of ~25,000 individuals representing nearly $1 million/yr for testing alone for a <0.5% positive test yield.
In summary, APAP toxicity/overdoses remain an ongoing problem in North America. However, many patients nowadays receive ED care precluding full hospitalization, and those requiring admission fare better than those of the earlier era with fewer deaths. Reasons for this improvement may include increased public awareness of APAP and/or the improvements in management protocols for the treatment of potential or actual APAP toxicity in the ED setting.
Financial support:
This work was supported by the Lawrence L. and Terry P. Tobin Fund for Liver Disease Research and the Jeanne Roberts Fund for Research and Treatment of Liver Disease, both at the Southwestern Medical Foundation, Dallas, TX.
Footnotes
Potential competing interests: None to report.
REFERENCES
- 1.Reuben A, Tillman H, Fontana RJ, et al. Outcomes in adults with acute liver failure between 1998 and 2013: An observational cohort study. Ann Intern Med 2016;164:724–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Schiødt FV, Rochling FA, Casey DL, et al. Acetaminophen toxicity in an urban county hospital. N Engl J Med 1997;337:1112–7. [DOI] [PubMed] [Google Scholar]
- 3.Larson AM, Polson J, Fontana RJ, et al. Acetaminophen-induced acute liver failure: Results of a United States multicenter, prospective study. Hepatology 2005;42:1364–72. [DOI] [PubMed] [Google Scholar]
- 4.Bateman DN, Vale A. Paracetamol (acetaminophen). Med (United Kingdom) 2016;375:2099. [Google Scholar]
- 5.Prescott LF, Critchley JAJH, Proudfoot AT, et al. Intravenous N-acetylcysteine: The treatment of choice for paracetamol poisoning. Br Med J 1979;2:1097–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rumack BH, Peterson RC, Koch GG, et al. Acetaminophen overdose: 662 cases with evaluation of oral acetylcysteine treatment. Arch Intern Med 1981;141:380–5. [DOI] [PubMed] [Google Scholar]
- 7.Caroll R, Benger J, Bramley K, et al. Epidemiology, management and outcome of paracetamol poisoning in an inner city emergency department. Emerg Med J 2015;32:155–60. [DOI] [PubMed] [Google Scholar]
- 8.Friðriksdóttir ÞA, Jónsdóttir F, Snook CP, et al. Paracetamol poisoning: A population-based study from Iceland. Scand J Gastroenterol 2021;56:832–9. [DOI] [PubMed] [Google Scholar]