Introduction
Ohio is experiencing unprecedented loss of life caused by unintentional drug overdoses,[1] with illicitly manufactured fentanyl (IMF) emerging as a significant threat to public health.[2,3] IMF is structurally similar to pharmaceutical fentanyl, but is produced in clandestine laboratories and includes fentanyl analogs that display wide variability in potency;[2] variations in chemical composition of these drugs make detection more difficult. During 2010–2015, unintentional drug overdose deaths in Ohio increased 98%, from 1,544 to 3,050.* In Montgomery County (county seat: Dayton), one of the epicenters of the opioid epidemic in the state, unintentional drug overdose deaths increased 40% in 1 year, from 249 in 2015 to 349 in 2016 (estimated unadjusted mortality rate = 57.7 per 100,000).[4] IMFs have not been part of routine toxicology testing at the coroner's offices and other types of medical and criminal justice settings across the country.[2,3] Thus, data on IMF test results in the current outbreak have been limited. The Wright State University and the Montgomery County Coroner's Office/Miami Valley Regional Crime Laboratory (MCCO/MVRCL) collaborated on a National Institutes of Health study of fentanyl analogs and metabolites and other drugs identified in 281 unintentional overdose fatalities in 24 Ohio counties during January–February 2017. Approximately 90% of all decedents tested positive for fentanyl, 48% for acryl fentanyl, 31% for furanyl fentanyl, and 8% for carfentanil. Pharmaceutical opioids were identified in 23% of cases, and heroin in 6%, with higher proportions of heroin-related deaths in Appalachian counties. The majority of decedents tested positive for more than one type of fentanyl. Evidence suggests the growing role of IMFs, and the declining presence of heroin and pharmaceutical opioids in unintentional overdose fatalities, compared with 2014–2016 data from Ohio and other states.[3–5] There is a need to include testing for IMFs as part of standard toxicology panels for biological specimens used in the medical, substance abuse treatment, and criminal justice settings.
The MCCO Toxicology laboratory provides postmortem forensic toxicology services to approximately 30 of Ohio's 88 counties. Data from 281 unintentional overdose fatalities that occurred in Montgomery County and 23 additional counties† during January and February 2017, were analyzed by the MCCO Toxicology laboratory, and had assigned causes of death as of May 8, 2017, were included in this study. Montgomery County data include all unintentional drug overdose deaths that occurred in the county during the specified period. Other county data include all cases that were sent to MCCO for analysis, but might not represent all unintentional overdose deaths that occurred in those counties. A liquid-chromatography-tandem mass spectrometry–based method, developed and validated by toxicologists at the MCCO Toxicology Laboratory and Department of Chemistry, Wright State University, was used to test for 25 fentanyl analogs, metabolites, and synthetic opioids§ in biologic matrices (human blood and urine specimens).
Toxicologic testing for other substances (heroin, pharmaceutical opioids, benzodiazepines, cocaine, methamphetamine, marijuana, and alcohol) was also conducted. Information on demographic characteristics including age, sex, and race was collected for each decedent. Counties were grouped into the following four urban/rural categories used by the Ohio Department of Health: 1) urban (Montgomery), 2) suburban, 3) rural, non-Appalachian, and 4) Appalachian. The chi-square statistic was used to assess differences among the four county groups in terms of demographic and drug-related characteristics. To examine polydrug patterns, reports of the presence of other fentanyl analogs/metabolites and other drugs were examined for decedents with positive test results for 1) fentanyl, 2) acryl fentanyl, 3) furanyl fentanyl, and 4) carfentanil, one of the most potent fentanyl analogs.
Among the 281 decedents, 122 (43.3%) were from Montgomery County (City of Dayton), a large urban county with a population of approximately 530,000 persons ( Table 1 ). Decedents from four suburban counties, who accounted for 52 (18.5%) unintentional overdose deaths, were primarily from areas that are a part of or adjacent to the Dayton Metro area. Seventy-six (27.0%) decedents were from rural, non-Appalachian counties, primarily from the Southwestern part of the state, and 31 (11.0%) were from the Appalachian counties that are located in the Southern part of the state.
Males accounted for 181 (64.4%) unintentional overdose deaths, and 257 (91.5%) decedents were white; this proportion was higher in rural (98.7%) and Appalachian (96.8%) counties (p = 0.007) ( Table 2 ). Over half (57.7%) of deaths occurred in persons aged 25–44 years. Approximately 7% of all decedents were not residents of the county where they died, with larger numbers of out of county resident deaths in urban Montgomery County (9.8%).
Overall, 253 (90.0%), 136 (48.4%), and 87 (31.0%) decedents tested positive for fentanyl, acryl fentanyl, and furanyl fentanyl, respectively ( Table 2 ). The proportions of decedents that were positive for acryl fentanyl and furanyl fentanyl were lower in Appalachian counties (29.0% and 19.4%, respectively), although these differences were not statistically significant. There were statistically significantly more decedents in urban and suburban counties that tested positive for despropionylfentanyl (4-ANPP) (45.1% and 55.8%, respectively) than in rural (34.2%) and Appalachian (25.6%) counties (p = 0.021).
Only 16 (5.7%) of all 281 decedents tested positive for heroin, with a significantly higher proportion in Appalachian counties (25.8%) than in urban (2.5%), suburban (3.8%) or rural non-Appalachian counties (3.9%). Among all 16 heroin-positive cases, 12 also tested positive for IMF. Overall, 64 (22.8%) decedents tested positive for pharmaceutical opioids, 75 (26.6%) for benzodiazepines, and 86 (30.6%) for cocaine; a higher percentage of decedents who tested positive for cocaine died in urban (37.7%) and suburban (42.3%) counties than in rural (22.4%) or Appalachian (3.2%) counties (p<0.001) ( Table 2 ).
Over half (53.8%) of specimens from fentanyl-positive decedents also tested positive for acryl fentanyl, and approximately one third (34.0%) for furanyl fentanyl ( Table 3 ). Approximately 62% of fentanyl-positive decedents did not test positive for norfentanyl. All specimens from acryl fentanyl deaths also tested positive for fentanyl, and 39.7% tested positive for furanyl fentanyl. Approximately 99% of furanyl fentanyl deaths tested positive for fentanyl, 62.1% for acryl fentanyl, and 86.2% for 4-ANPP.
Twenty-one decedents (including 11 [52%] in Montgomery County) tested positive for carfentanil. Among these, 15 (71.4%) decedents also tested positive for fentanyl, five (23.8%) for acryl fentanyl, and eight (38.1%) for furanyl fentanyl. Many of the carfentanil decedents tested positive for other central nervous system depressants, such as pharmaceutical opioids (23.8%) and benzodiazepines (42.9%). Approximately 30% of fentanyl, acryl fentanyl, and furanyl fentanyl cases tested positive for cocaine. Among carfentanil cases, approximately 40% were positive for cocaine ( Table 3 ).
Morbidity and Mortality Weekly Report. 2017;66(34):904-908. © 2017 Centers for Disease Control and Prevention (CDC)
