Delayed-type hypersensitivity reactions induced by proton pump inhibitors: A clinical and in vitro T-cell reactivity study.

BACKGROUND: Proton pump inhibitors (PPIs) have been known to induce type I hypersensitivity reactions. However, severe delayed-type hypersensitivity reactions (DHR) induced by PPI, such as Stevens-Johnson syndrome (SJS), toxic epidermal necrolysis (TEN), or drug rash with eosinophilia and systemic symptoms (DRESS), are rarely reported. We conducted a study of a large series of PPI-related DHR, followed up their tolerability to alternative anti-ulcer agents, and investigated the T-cell reactivity to PPI in PPI-related DHR patients. METHODS: We retrospectively analyzed patients with PPI-related DHR from multiple medical centers in Taiwan during the study period January 2003 to April 2016. We analyzed the causative PPI, clinical manifestations, organ involvement, treatment, and complications. We also followed up the potential risk of cross-hypersensitivity or tolerability to other PPI after their hypersensitivity episodes. Drug lymphocyte activation test (LAT) was conducted by measuring granulysin and interferon-γ to confirm the causalities. RESULTS: There were 69 cases of PPI-related DHR, including SJS/TEN (n=27) and DRESS (n=10). The LAT by measuring granulysin showed a sensitivity of 59.3% and specificity of 96.4%. Esomeprazole was the most commonly involved in PPI-related DHR (51%). Thirteen patients allergic to one kind of PPI could tolerate other structurally different PPI without cross-hypersensitivity reactions, whereas three patients developed cross-hypersensitivity reactions to alternative structurally similar PPI. The cross-reactivity to structurally similar PPI was also observed in LAT assay. CONCLUSIONS: PPIs have the potential to induce life-threatening DHR. In patients when PPI is necessary for treatment, switching to structurally different alternatives should be considered.

Delayed hypersensitivity on a surgical service.

Two hundred patients assessed for nutritional deficiencies were analyzed for reaction to skin testing with common antigens and the incidence of sepsis and death. Only 39% of all skin tests were positive, though 50% of the patients had at least one positive test. Associated diagnoses revealed a high incidence of malnutrition, cancer, radiation therapy, and chemotherapy. Analysis using Chi-square and Gamma shows good statistical correlation between skin tests and sepsis and death. Those with negative tests (anergic) using PPD, Candida, and mumps had a threefold higher mortality. Major sepsis also increased in the skin negative group (+80%), but here the incidence varied directly with the number of positive skin tests. Mortality, unlike sepsis, was influenced only by the presence of delayed hypersensitivity and was not related to the number of positive reactions. The basic rate of infection or mortality was not influenced by major surgery.

[Nutritional status in critically ill patients. Relationship with mortality (author's transl)]. / Etat nutritionnel des malades admis en réanimation. Relation avec la mortalité.

Plasma proteins, triglyceridemia, body composition and delayed hypersensitivity were determined in 154 critically ill patients after admission. Plasma proteins levels were significantly increased in patients that were subsequently discharged vs. those that died: albumin: 33 +/- 6 g/l vs 28 +/- 6 g/l (p < 10(-6)); transferrin 2,18 +/- 0,65 g/l vs. 1,54 +/0 0,55 g/l (p < 10(-7)); prealbumin: 14,32 +/- 7,79 mg/100 ml vs. 7,28 +/-5,36 mg/100 ml (p < 10(-7)) and triglyceridemia was decreased: 1,07 +/- 0,38 g/l vs. 1,66 +/- 1,12 g/l (p not equal to 10(-3)). Body weight, fat weight and lead body mass were not correlated to subsequent mortality. Muscle cell mass was decreased (-17%, p < 10(-2)) and extracellular water was increased (+14%, p < 10(-4)), in patients who subsequently died. Total body water and visceral cell mass did not change. Initial anergy (tested with 3 antigens: candidin, tuberculin, varidase) did correlate with mortality: 35/62 died when delayed hypersensitivity was negative vs. 13/71 when it was positive (p < 10(-4)). Mortality was associated with decreased total lymphocyte count: 884 +/- 1025 vs. 1270 +/- 870 (p < 0,02) and serum iron: 51 +/- 40 micrograms/100 ml vs. 74 +/- 45 micrograms/100 ml (p < 10(-2)). Sepsis correlated with mortality (p < 10(-3)) and could produce these changes. These results suggest that critically ill paients have a protein-calorie malnutrition syndrom marktly different from that observed in simple starvation. Nutritional therapy must be, in this group of patients, adapted to this concept.