Estudio observacional retrospectivo sobre interacciones farmacológicas en la fase de post-trasplante en pacientes sometidos a trasplante de progenitores hematopoyéticos

Resumo

Patients undergoing hematopoietic stem cell transplantation (HSCT) are treated with highly complex drug regimens, the combination of which can lead to multiple pharmacological interactions (PI), which may increase the risk of adverse effects or reduce the effectiveness of the treatment. Knowledge and proper management of PIs can contribute to improving the quality, safety and efficiency of the treatments administered to patients. There are few studies of PIs in patients undergoing HSCT, and furthermore, there are fewer studies focused on the post-transplant stage, so the incidence of interactions, their potential severity and risk reduction strategies are poorly defined in this context. For all the above, the objectives of this study were to determine the prevalence and frequency of potential contraindicated and major PIs, in the immediate post-transplant phase, in patients undergoing HSCT, using two databases, Micromedex¿ and Lexicomp¿; and the risk factors associated with the presence of contraindicated and major interactions in this group of patients; as well as describing the most frequent and potentially more serious interactions and proposing pharmacological alternatives or risk reduction strategies, prioritizing the most frequent contraindicated and major interactions. Secondarily, it was proposed to carry out a simulation of the potential contraindicated and major PIs that the patients would have had if the proposed alternatives had been prescribed instead of the original drugs and to analyze the interactions obtained in the simulation, checking if the frequency of appearance of PIs was reduced. A descriptive, retrospective, observational study was designed, including all patients over 16 years of age in charge of the Hematology Service who underwent HSCT, between the 1st of January of 2016 and the 31st of December of 2018. Data were collected on the characteristics of the patients, the characteristics of the disease and the prescribed treatment. When a patient underwent more than one transplant during the study period, each episode was treated as a separate patient. The treatments administered from day +1 (day after the infusion of progenitors) and the day of the neutrophil graft were evaluated, with a maximum of 21 days. All support and prophylaxis drugs for graft versus receptor disease (GVRD) were recorded, as well as the rest of the drugs prescribed in the immediate post-transplant stage. For the analysis of the treatments, two databases were used, Micromedex¿ and Lexicomp¿, recording the contraindicated and major potential PIs detected (according to Micromedex¿) and those of risk level X and D (according to Lexicomp¿), as well as the level of severity, the level of evidence, the mechanism of the PIs, its description and the duration of the PIs. A total of 140 treatments were analyzed, belonging to 131 patients, with a median (range) age of 57 (16-73) years. Eighty-five percent of the patients had an ECOG between 0 and 1, and the most frequent diagnosis was multiple myeloma (43.6%). The majority (81.4%) of the patients had not received any previous HSCT and more than a half (50.7%) of the patients had a complete response before undergoing HSCT. Autologous transplantation was the most common (71.43%), the most frequent source of progenitors was peripheral blood (97.9%), and the predominant intensity of conditioning was myeloablative (79.3%). The mean number of days of treatment collected was 12.67 ± 3.72 days. A total of 3,622 drugs were prescribed (1,893 for support, 105 for the prophylaxis of GVRD and 1,624 for the rest of the drugs). A total of 1,956 PIs (401 drug pairs) were detected by the Micromedex¿ database, 251 PIs of which (25 pairs) were categorized as contraindicated and 1,705 PIs (376 pairs) as major; and a total of 1,008 PIs (240 pairs) were detected by the Lexicomp¿ database, 92 PIs of which (26 pairs) were classified as risk level X and 916 PIs (214 pairs) as level D. The prevalence of potential total PIs was 100% for Micromedex¿ (95.71% for contraindicated PIs and 99.29% for severe PIs) and the prevalence of total potential PIs for Lexicomp¿ was 92.14% (40% for level X and 91.43% for those of level D). The drugs involved in the highest number of PIs were those of the group "other drugs" and the predominant mechanism of the PIs was pharmacodynamic. Some PIs stood out due to their frequency and potential severity in the study, such as interactions that implied a risk of prolongation of the QT Interval, being the drugs most frequently implicated azoles, calcineurin inhibitors and ondansetron; interactions that increase the risk of extrapyramidal reactions, with metoclopramide as the main agent involved; and those interactions associated with a risk of central nervous system depression, being the drugs associated with the highest risk opioids, benzodiazepines, and metoclopramide. In the multivariate analysis, the probability of patients suffering PIs, in both databases, was associated with the number of total drugs prescribed. The main alternatives identified with lower risk of potential significant interactions were cyclosporine A instead of tacrolimus, mycophenolate sodium instead of mycophenolate mofetil, micafungin instead of fluconazole, morphine replacing fentanyl, avoiding combinations of codeine with morphine, tramadol or paracetamol replacing metamizole, duloxetine as a possible antidepressant of choice, and chlorthalidone instead of furosemide. The prevalence of potential contraindicated and major PIs, obtained after performing the simulation with the proposed changes, decreased in a statistically significant way, in both databases, especially in the case of the Lexicomp¿ database, in which the prevalence of risk level X PIs was reduced by almost 50%.