Русский
Adverse reactions in children to antimicrobials: limitations of the spontaneous reporting method and the possibilities of the global trigger method for drug-induced conditions
DOI: https://dx.doi.org/10.18565/pharmateca.2023.1-2.18-31
A.V. Vlasova, E.V. Smirnova, V.V. Gorev, D.A. Sychev
1) Morozov Children’s City Clinical Hospital, Moscow, Russia; 2) Russian Medical Academy of Continuous Professional Education, Moscow, Russia
Backgroun. The safety of pharmacotherapy in critically ill children with nosocomial infection is an urgent problem in pediatrics [1]. The main cause of adverse drug reactions (ADRs) in children is antimicrobial drugs (AMDs) [2, 3]. The inefficiency of the traditional method of retrospective registration of spontaneous reports (SR) of ADRs was shown. The theoretical foundations for the introduction of the method for assessing global triggers (GT) of drug-induced conditions as a transition to the tactics of active detection of ADRs are given [4–10].
Objective. Comparative analysis of the prevalence and structure of ADRs to AMDs in children when registering drug-induced conditions using the SR and GT methods.
Methods. A comparative analysis of the prevalence and structure of manifestations of ADRs to AMDs in children was performed when registered by the SR method retrospectively for those hospitalized in the pediatrics profile, incl. in critical conditions during hospitalization in the ICU on the profile «anesthesiology-resuscitation (children)» for 3 years and with active tactics for detecting drug-induced conditions on AMDs in the ICU by the GT method for 2 years.
Results. The retrospective prevalence of ADRs to AMDs was 0.32 (95% CI: 0.22–0.33) per 100 hospitalized pediatric patients, the structure of ADR manifestations in 40 cases: allergic reactions – in 25 (62%) cases (cephalosporins and vancomycin), no expected therapeutic effect – in 7 (18%) cases (ceftriaxone, ampicillin sulbactam, azithromycin, levofloxacin, imipenem cilastatin, tobramycin inhalation, linezolid), vomiting and diarrhea – in 3 (7%) (amoxicillin clavulanate), bronchospasm – in 2 (5%) (colistimethate inhalation use), agitation – in 1 (3%) (ceftriaxone), nephrotoxicity – in 1 (3%) (amikacin), and phlebitis – in 1 (3%) case (vancomycin). For the SR method, the accuracy rate was 60% (95% CI: 54.69–65.03), sensitivity was 90.9% (95% CI: 78.33–97.47), and specificity was 55.7% (95 % CI: 50.01–61.20). In children in the ICU using the SR method, no ADR on AMD was registered. In a prospective observational study using the GT assessment method, the prevalence of manifestations of ADRs on AMDs in critically ill children was 1.28 (95% CI: 1.22–1.36) per 100 hospitalized in the ICU. The structure of ADRs and suspected drugs for 30 children were represented by hepatotoxicity – 19 (63%; cefoperazone sulbactam, vancomycin, tigecycline, meropenem, fluconazole, aztreonam, ceftazidime-avibactam), nephrotoxicity – 3 (10%; vancomycin), cardiovascular manifestations – 2 (7%; ciprofloxacin, moxifloxacin), neurotoxicity – 2 (7%; colistimethate, posaconazole), gastroenterological manifestations – 2 (7%) and laboratory abnormalities (vancomycin) – 2 (7%) children. For the method of active detection of ADRs to AMDs in the assessment of GT drug-induced conditions, the accuracy rate was 97% (95% CI: 95.86–97.37), the sensitivity of the method was 81.1% (95% CI: 64.84–92.04); specificity –
97.0% (95% CI: 96.14–97.60).
Results. Serious limitations of the SR method for registering ADRs for AMDs in a pediatric hospital in real clinical practice for 3 years are shown. The prevalence of ADRs by SR was 0.32 per 100 hospitalized. Its inefficiency for children in critical conditions was shown on the example of the ICU of a multidisciplinary children’s hospital. The prevalence of ADRs on AMDs in critically ill children with the use of active tactics for detecting drug-induced conditions by GT for 2 years was significantly higher and amounted to 1.28 per 100 hospitalized in the ICU. The accuracy rate of the method of active detection of ADRs on AMDs was 97%, the method of GT of drug-induced conditions – 60%.In the present study of critically ill children with infection, the risk of developing a fatal outcome was 6 times higher in the group of patients with manifestations of ADRs on AMDs compared with patients who did not have ADRs with active detection by GT (OR=6.0; 95% CI: 2.06–17.48).
Conclusion. The accuracy rate of the method of active detection of ADRs to AMDs when assessing drug-induced conditions using the GT method was 97% and was higher compared to the accuracy rate of the retrospective detection of ADRs to AMDs using the SR method, which was 60%. For the first time in the present study, the risk of developing a lethal outcome was 6 times higher in the group of patients with manifestations of ADRs to AMDs compared with patients who did not have ADRs with active tactics of detection by GT (OR=6.0; 95% CI: 2.06–17,48) in critically ill children with infection.
Objective. Comparative analysis of the prevalence and structure of ADRs to AMDs in children when registering drug-induced conditions using the SR and GT methods.
Methods. A comparative analysis of the prevalence and structure of manifestations of ADRs to AMDs in children was performed when registered by the SR method retrospectively for those hospitalized in the pediatrics profile, incl. in critical conditions during hospitalization in the ICU on the profile «anesthesiology-resuscitation (children)» for 3 years and with active tactics for detecting drug-induced conditions on AMDs in the ICU by the GT method for 2 years.
Results. The retrospective prevalence of ADRs to AMDs was 0.32 (95% CI: 0.22–0.33) per 100 hospitalized pediatric patients, the structure of ADR manifestations in 40 cases: allergic reactions – in 25 (62%) cases (cephalosporins and vancomycin), no expected therapeutic effect – in 7 (18%) cases (ceftriaxone, ampicillin sulbactam, azithromycin, levofloxacin, imipenem cilastatin, tobramycin inhalation, linezolid), vomiting and diarrhea – in 3 (7%) (amoxicillin clavulanate), bronchospasm – in 2 (5%) (colistimethate inhalation use), agitation – in 1 (3%) (ceftriaxone), nephrotoxicity – in 1 (3%) (amikacin), and phlebitis – in 1 (3%) case (vancomycin). For the SR method, the accuracy rate was 60% (95% CI: 54.69–65.03), sensitivity was 90.9% (95% CI: 78.33–97.47), and specificity was 55.7% (95 % CI: 50.01–61.20). In children in the ICU using the SR method, no ADR on AMD was registered. In a prospective observational study using the GT assessment method, the prevalence of manifestations of ADRs on AMDs in critically ill children was 1.28 (95% CI: 1.22–1.36) per 100 hospitalized in the ICU. The structure of ADRs and suspected drugs for 30 children were represented by hepatotoxicity – 19 (63%; cefoperazone sulbactam, vancomycin, tigecycline, meropenem, fluconazole, aztreonam, ceftazidime-avibactam), nephrotoxicity – 3 (10%; vancomycin), cardiovascular manifestations – 2 (7%; ciprofloxacin, moxifloxacin), neurotoxicity – 2 (7%; colistimethate, posaconazole), gastroenterological manifestations – 2 (7%) and laboratory abnormalities (vancomycin) – 2 (7%) children. For the method of active detection of ADRs to AMDs in the assessment of GT drug-induced conditions, the accuracy rate was 97% (95% CI: 95.86–97.37), the sensitivity of the method was 81.1% (95% CI: 64.84–92.04); specificity –
97.0% (95% CI: 96.14–97.60).
Results. Serious limitations of the SR method for registering ADRs for AMDs in a pediatric hospital in real clinical practice for 3 years are shown. The prevalence of ADRs by SR was 0.32 per 100 hospitalized. Its inefficiency for children in critical conditions was shown on the example of the ICU of a multidisciplinary children’s hospital. The prevalence of ADRs on AMDs in critically ill children with the use of active tactics for detecting drug-induced conditions by GT for 2 years was significantly higher and amounted to 1.28 per 100 hospitalized in the ICU. The accuracy rate of the method of active detection of ADRs on AMDs was 97%, the method of GT of drug-induced conditions – 60%.In the present study of critically ill children with infection, the risk of developing a fatal outcome was 6 times higher in the group of patients with manifestations of ADRs on AMDs compared with patients who did not have ADRs with active detection by GT (OR=6.0; 95% CI: 2.06–17.48).
Conclusion. The accuracy rate of the method of active detection of ADRs to AMDs when assessing drug-induced conditions using the GT method was 97% and was higher compared to the accuracy rate of the retrospective detection of ADRs to AMDs using the SR method, which was 60%. For the first time in the present study, the risk of developing a lethal outcome was 6 times higher in the group of patients with manifestations of ADRs to AMDs compared with patients who did not have ADRs with active tactics of detection by GT (OR=6.0; 95% CI: 2.06–17,48) in critically ill children with infection.
About the Autors
Corresponding author: Anna V. Vlasova, Cand. Sci. (Med.), Head of the Department of Clinical Pharmacology, Morozov Children’s City Clinical Hospital; Associate Professor at the Department of Clinical Pharmacology and Therapy named after Acad. B.E. Votchal, Russian Medical Academy of Continuous Professional Education, Moscow, Russia; annavlasova75@mail.ru
Similar Articles
