On July 10, 2019, Wisconsin Department of Health Services (WDHS) was notified of five previously healthy adolescents with severe lung injuries who reported use of e-cigarette, or vaping, products before symptom onset. As of December 31, 2019, 105 confirmed or probable cases of e-cigarette, or vaping, product use–associated lung injury (EVALI)* had been reported to WDHS . Three social clusters (A, B, and C), comprising eight EVALI patients (cluster A = two patients, cluster B = three, and cluster C = three) were identified. WDHS investigated these clusters with standard and follow-up interviews; laboratory analysis of e-cigarette, or vaping, products; and analysis of bronchoalveolar lavage (BAL) fluid. All eight patients reported daily use of tetrahydrocannabinol (THC)-containing e-cigarette, or vaping, product cartridges (THC cartridges) in the month preceding symptom onset. All THC cartridges were purchased from local illicit dealers, and all patients reported using THC cartridges labeled as “Dank Vapes,” among other illicit brand names. At least two members of each cluster reported frequent sharing of THC cartridges before symptom onset. All eight patients also reported daily use of nicotine-containing e-cigarette, or vaping, products. Vitamin E acetate (VEA) was detected in all five THC cartridges tested from two patients, and in BAL fluid from two other patients. These findings suggest that THC cartridges containing VEA and sold on the illicit market were likely responsible for these small clusters of EVALI. Based on information presented in this and previous reports (1,2) CDC recommends not using THC-containing e-cigarette, or vaping, products, especially those obtained from informal sources such as friends, family, or in-person or online dealers (1). VEA is strongly linked to the EVALI outbreak and should not be added to e-cigarette, or vaping, products (1).

A cluster was defined as two or more patients with confirmed or probable EVALI who directly shared e-cigarette, or vaping, products; obtained products from the same source; or reported a social connection and use of the same e-cigarette, or vaping, product brand names in the 3 months preceding symptom onset. All patients were interviewed by telephone using a standard EVALI questionnaire developed by WDHS, and five of the eight cluster-associated patients (A = one, B = three, C = one) completed in-depth follow-up telephone interviews to provide additional product use details. This included, for each product used, the dates of initiation and cessation, frequency and amount used, and the extent of sharing with other EVALI patients. In addition to interviews, one patient in cluster A and two patients in cluster C submitted a total of 11 e-cigarette, or vaping, products that were tested for the presence of VEA and other additives^† by the Food and Drug Administration (FDA), and BAL fluid from two patients (one each from cluster B and cluster C) were analyzed by CDC (3).^§

Symptom onset for these eight patients ranged from June 18 through July 21, 2019 (Figure 1). Patients were aged 16–20 years (median = 17 years), and six were male. All eight patients reported daily use of THC cartridges purchased from local illicit dealers in the month before symptom onset. This included use of the Dank Vapes brand by all patients and an average of 2.6 unique brands of illicit THC cartridges per patient (range = one to five brands) (Table). At least two patients in each cluster reported frequent sharing of THC cartridges in the month preceding symptom onset, including concurrent use of the same cartridge in the same device (Table). On average, patients reported inhaling approximately one half of a 1-g THC cartridge daily (range = 0.2–1 cartridge per day) in the month before symptom onset; two patients (one in cluster B and one in cluster C) reported that this was more than usual for them. All patients also reported daily use of nicotine-containing e-cigarette, or vaping, products. These included commercial pods and refillable e-liquids purchased from retail locations or online. The amount of nicotine product use per day was not quantifiable because of variability among brands. Patients reported initiating use of THC cartridges a median of 9 months before onset of symptoms (range = <1 to 12 months) (Figure 2). Patients in cluster A initiated daily use of Dank Vapes 2–4 weeks before symptom onset, whereas patients in clusters B and C reported a longer duration of THC cartridge use before symptom onset, without changing brands or sources. All patients reported long-term use of nicotine-containing products, which were initiated a median of 33 months before symptom onset (range = 5–60 months) (Figure 2).

Eleven e-cigarette, or vaping, products from three patients were tested. All five THC cartridges collected from two patients contained VEA; one product contained nicotine, VEA, and cannabinol; none of the five commercial nicotine products collected from two patients contained VEA. None of the products tested contained significant levels of other toxicants included in the FDA testing protocol. BAL fluids were tested for two patients, and both contained VEA; no other potential toxicants were identified in these BAL fluids.

Injury severity and clinical course varied among these eight patients (Table). Six patients were hospitalized for a median of 6.5 days (range = 6–20), five were admitted to the intensive care unit, and two required mechanical ventilation. Two patients from cluster A received a diagnosis of EVALI in outpatient settings. One patient from cluster B reported persistent respiratory symptoms 3 months after discharge.