COMUNICACIÓN BREVE

 

Is there an overprescription of proton pump inhibitors in oncohematologic patients undergoing ambulatory oncospecific treatment?

¿Existe una sobreprescripción de inhibidores de la bomba de protones en el paciente oncohematológico en tratamiento oncoespecífico ambulatorio?

 

 

Meritxell Pujal Herranz

Pharmacy Unit. Hospital de Terrassa. Consorci Sanitari de Terrassa, Barcelona. Spain.

Correspondence

 

 

---

ABSTRACT

Objective: The aim of this study is to evaluate the prevalence of proton pump inhibitors (PPIs) prescription, and the level of adequacy of the indication of these drugs in oncohematologic patients under ambulatory oncoespecific treatment.
Method: An observational descriptive study in oncohematologic patients under ambulatory oncoespecific treatment. A protocol for the rational use of PPI targeted to oncohematologic patients based on the PPI protocol of our hospital was designed. Patients under active treatment with PPIs were quantified and the appropriateness of their indications evaluated.
Results: 111 patients (71 oncologic and 40 hematologic) were included. 56% of all oncologic patients and 63% of all hematologic patients were under active treatment with PPIs. After reviewing the indications for PPI in all patients, 72% of oncologic and 12% of hematologic patients did not present evidence justifying treatment with these drugs.
Conclusion: It is important the pharmacist to detect unappropriated prescriptions of PPIs, especially among oncologic patients, and to promote a deprescription of these drugs.

Key words: Proton pump inhibitors; Oncologic patients; Inappropriate prescription; Pharmaceutical care.

---

RESUMEN

Objetivo: El objetivo de este estudio es analizar la prevalencia de inhibidores de la bomba de protones (IBP) en el paciente oncohematológico de dispensación ambulatoria y el grado de adecuación de su indicación.
Método: Estudio observacional descriptivo en pacientes oncohematológicos en tratamiento oncoespecífico de dispensación ambulatoria. Se elaboró un protocolo dirigido al paciente oncohematológico a partir del protocolo de uso racional de IBP de nuestro hospital. Se cuantificaron los pacientes en tratamiento activo con IBP y se analizó la idoneidad de su indicación.
Resultados: Se incluyeron 111 pacientes (71 oncológicos, 40 hematológicos). El 56% de los pacientes oncológicos y el 63% de los hematológicos estaban en tratamiento activo con IBP. Tras revisar las indicaciones de los pacientes con IBP, el 72% de los oncológicos y el 12% de los hematológicos no presentaron una indicación que justificara el tratamiento.
Conclusiones: Es importante que el farmacéutico detecte las prescripciones inadecuadas de IBP especialmente entre la población oncológica y sugiera una deprescripción del mismo.

Palabras clave: Inhibidor de la bomba de protones; Paciente oncológico; Sobreprescripción; Intervención farmacéutica.

---

 

Contribution to scientific literature

This study presents data, so far unknown, on the prevalence of prescription of proton pump inhibitors (PPIs) and the level of adequacy of their indication in oncohematologic patients undergoing ambulatory oncospecific treatment.

The results of this study reflect the need to review the indication for PPIs on oncohematologic patients under ambulatory oncospecific treatment. The aim is to simplify treatment in those patients for whom PPI prescription is not justified, in order to minimize as much as possible the potential side effects derived of these drugs.

 

Introduction

There is evidence about overprescription of proton pump inhibitor drugs (PPIs) in various studies, both at Primary Care level and in the hospital setting. A recent published study involving 112 patients observed that 69% of them took a PPI as regular medication at the time of admission. After analyzing the indication for PPI in these patients, it was demonstrated that only 24% of cases presented an adequate indication¹. Moreover, there are studies showing that 33% of those patients who initiate treatment with a PPI due to a sporadic clinical prescription, will continue taking it after the required period of time, without an obvious indication of treatment^2-3. These data suggest that there is a high proportion of patients on treatment with a PPI with an unappropriated prescription among the general population.

Some of the likely causes for the high use of PPIs are: the false belief that these are harmless medications, population ageing, self-medication, an unappropriated indication for gastric protection with non-steroid anti-inflammatory drugs (NSAIDs) without any other risk factors, their frequent prescription as empiric treatment for dyspepsia or minor gastric conditions, and the lack of periodical reviews of treatments⁴.

The therapeutical indications of PPIs are: gastric protection in chronic treatments with two concomitant drugs with high gastrolesivity, or one single drug in patients with risk factors, and the treatment of digestive conditions such as gastroesophageal reflux disease, Barrett's oesophagus, peptic ulcers, eradication of Helicobacter pylori, and non-ulcerous dyspepsia. For oncohematologic patients, there is only one additional indication to those previously mentioned. American Clinical Guidelines suggest the potential use of PPIs as coadjuvant treatment for chemotherapy-induced nausea and/or vomiting in patients with dyspepsia⁵.

PPIs, when prescribed according to their indication and for an adequate duration, are overall well tolerated, with some mild adverse effects observed in clinical trials. However, they can cause safety problems, mostly in prolonged treatments. In recent years, various articles have associated them with adverse effects that, even though not frequent, can be potentially serious, such as hypomagnesemia, and an increase in the risk of bone fractures, pneumonias, and intestinal infections by Clostridium difficile^6-10.

In 2011, the Spanish Agency of Medicines and Medical Devices issued an informative note associating the use of PPIs with cases of severe hypomagnesemia⁶. Even though the mechanism is unknown, it is thought that the prolonged use of a PPI can interfere in the intestinal absorption of magnesium, leading to hypomagnesemia which, at the same time, can cause hypoparathyroidism, hypocalcemia, and hypopotasemia. The risk of hypomagnesemia can be increased in diabetic patients on prolonged treatments with a PPI, or those on concomitant treatments with digoxin or drugs that can cause hypomagnesemia on their own, such as thiazide diuretics and other drugs used for oncohematologic patients (cisplatin, carboplatin, cetuximab, panitumumab, decitabine, arsenic trioxide, aldesleukin, and cyclosporine)^6,7,11. For this reason, the oncohematologic population on treatment with a PPI in combination with any of these drugs will present an additional risk factor for developing hypomagnesemia.

PPIs can also increase the risk of hip, wrist and vertebral fractures, when taken at high doses for over one year of treatment, mostly in elderly patients, or when other risk factors for fracture are also present⁸. Osteoporosis secondary to bone metastasis presented by some oncohematologic patients could represent an additional risk factor in this population.

There is one meta-analysis establishing an association between the use of PPI and the development of community-acquired pneumonia and nosocomial pneumonia⁹. The sustained increase in gastric pH caused by a PPI has also been associated with a higher overgrowth of gastrointestinal bacterial flora, increasing the risk of infection by Clostridium difficile¹⁰. The presence of temporary immunosuppression secondary to oncospecific treatment or due to the oncologic disease in some oncohematologic patients could represent a risk factor for developing these infections.

Even though the incidence of these adverse effects is low, so far there is no accurate data of their frequency in the general population or, more specifically, in the oncohematologic population, as well as the consequences they might have on the quality of life of these patients. For this reason, it is important to report these cases to the Pharmacovigilance Centre, even those adverse effects already described.

The importance of obtaining data about the level of adequacy for PPI prescription in the oncohematologic population lies in the fact that this is a population that frequently presents intrinsic risk factors for developing the adverse effects previously described.

On the other hand, the potential drug interactions between PPIs and oral drugs with absorption depending on the gastric pH should not be underestimated. Some drugs frequently used for oncohematologic patients could present a reduced absorption when used in combination with a PPI. Some of them are: iron and calcium salts, antifungals, and oncospecific treatments with oral administration, such as certain tyrosine kinase inhibitors (erlotinib, gefitinib, lapatinib and dasatinib)¹². There is a current warning against the joint administration of a PPI with erlotinib, gefitinib or lapatinib, and there is a complete contraindication in the case of dasatinib. There are also interactions at the level of liver elimination, such as the association of a PPI with anagrelide or methotrexate at high doses. The mechanism of interaction is caused by a weak inhibition by the PPI of the CYP1A2 isoenzyme, which is in charge of eliminating both drugs. For this reason, there is also a warning against the use of this combination.

The objective of this study is to evaluate the prevalence of PPI prescription, and the level of adequacy of its indication for oncohematologic patients under ambulatory oncospecific treatment.

 

Methods

An observational descriptive study of the prevalence of the use of PPI, and the level of adequacy of the indication for these drugs in oncologic and hematologic patients under ambulatory oncospecific treatment.

The study included those patients who were on oncospecific treatment with ambulatory dispensing at the start of the recruitment period (December, 2014) and those who initiated that treatment within the following 6 months (from December, 2014 to May, 2015). Patients included in the study needed to be under oncospecific treatment with oral drugs or, in some cases, with subcutaneous drugs such as erythropoietins indicated for low-grade myelodysplastic syndrome. The study excluded those patients under oncospecific treatment with exclusive intravenous administration at Outpatient Hospital, because there were no data on initial and follow-up visits to the Pharmacy Unit which allowed to conduct an appropriated clinical interview.

The regular pharmacological treatment of these patients was reviewed through the electronic prescription record, in order to identify any existing concomitant treatment with a PPI. Confirmation that they were taking this medication was obtained through a clinical interview with the patient.

A protocol for the rational use of PPI targeted to oncohematologic patients was designed based on the PPI protocol of our hospital in consensus with the Oncology and Haematology Unit, according to Clinical Oncology Guidelines⁵. This protocol contained the recommendations for use of PPIs in oncohematologic patients (Table 1).

The proportion of patients on active treatment with a PPI, relative to the total population analyzed in the study was estimated. Also, the adequacy for PPI indication in the group of patients on active treatment was reviewed, based on the previously mentioned protocol for the rational use of PPIs in oncohematologic patients (Table 1), the clinical interview with the patient at the Pharmacy Unit, a complete review of their clinical history and, in case of doubt, a consultation with the specialist or primary care physician responsible for the patient.

Separate analyses for the oncologic and hematologic populations was decided before initiating patient inclusion in the study, in order to prevent a possible confusion bias caused by the difference in the typical mean age of both populations. This age difference could generate a confusion bias in the analysis of outcomes, because patients older than 70-year-old are more frequent among the hematologic population, and this risk factor would justify gastric protection with a PPI in chronic treatments with a highly gastrolesive drug (Table 1). At the same time, use of highly gastrolesive drugs such as antiaggregants and anticoagulants is also more frequent among the advanced-age population, because these patients present a higher incidence of cardiovascular conditions.

 

Outcomes

The study included 111 patients under ambulatory oncospecific treatment: 71 oncology patients (44 men and 27 women) with a mean age (SD) of 62 (13) years, and 40 hematology patients (28 men and 12 women) with a mean age (SD) of 71 (17) years. The type of tumour and the oncospecific treatment received by the study population appear in Table 2. During the period of the study, melphalan was dispensed at hospital as an exception, due to a temporary shortage in pharmacies. A 23% of the oncology patients in the study presented bone metastasis, and 18% of the hematology patients presented osteolytic lesions in the setting of multiple myeloma.

The proportion of patients in the study on treatment with a PPI was 56% in the oncologic population (40 patients, 23 men and 17 women) and 63% in the hematologic population (25 patients, 17 men and 8 women). The mean age (SD) of the oncologic patients on treatment with a PPI was 65 (13) years, while in hematologic patients it was 75 (9) years. After reviewing the indication for PPI in patients under active treatment with this drug (n=65, 40 oncologic and 25 hematologic patients), it was observed that 88% (22/25) of the hematologic population presented an appropriate indication, while this only happened in 28% (11/40) of the oncologic population (Fig. 1).

The indications for treatment with a PPI analyzed in the hematologic population were: gastric protection in 72% of cases, digestive conditions in 14%, and both indications simultaneously in another 14%. All the hematologic patients in the study on PPI and indication for gastric protection were on chronic treatment with one or more high gastrolesive drugs (ASA, corticosteroids and/ or heparin), and were >70-year-old. The digestive conditions registered in the hematological population that justified treatment with a PPI were: gastroesophageal reflux disease, peptic ulcers, and non-ulcerous dyspepsia.

Among the oncologic population, the indications for treatment with a PPI were gastric protection in 91% of cases, and digestive conditions in the remaining 9%. The oncologic patients with PPIs for gastric protection were on chronic treatment with a highly gastrolesive drug (ASA, clopidogrel, heparin, acenocoumarol, corticosteroids) and were older than 70-year-old; or were on chronic treatment with two highly gastrolesive drugs (corticosteroids + NSAIDs or corticosteroids + ASA), or with a highly gastrolesive drug in concomitance with a weak gastrolesive drug (heparin + citalopram). Regarding the indication of PPI for digestive conditions, only one patient with non-ulcerous dyspepsia was reported.

No patients in the study who took PPIs for dyspepsia as coadjuvant treatment for chemotherapy-induced nausea and/or vomiting were reported.

 

Discussion

The data obtained in the study on the prevalence of PPI prescription in the oncologic and hematologic population, of 56% and 63% respectively, are similar to those published so far for the general population¹.

The analysis for PPI indications in patients on IPP active treatment suggests that the proportion of patients on treatment with PPI with an adequate indication is high among the hematologic population, with an estimated 88%. This is in contrast to the oncologic population, for whom the use of PPIs is justified only in 28% of patients. There is a low proportion of oncologic patients in the study on treatment with a PPI with clinical indication. Similar results have been previously published concerning use of PPI within general population¹. The advanced age in the hematologic population, and the higher incidence of cardiovascular conditions requiring highly gastrolesive drugs, justifies the use of PPIs in a higher proportion of patients as in comparison with the oncologic population.

It is worth pointing out that no patients in the study on treatment with a PPI presented the indication of coadjuvant treatment for dyspepsia associated to chemotherapy-induced nausea and/or vomiting. The patients in the study underwent oncospecific treatment with moderate (vinorelbine and imatinib), low and minimal emetogenic power, and did not require any coadjuvant treatments for management of chemotherapy-induced nausea and/or vomiting.

Regarding the importance of the potential drug interactions of PPIs with any oral oncospecific treatments, 4 patients in the study (three of them on treatment with erlotinib and one with dasatinib) were found to have a potential risk of loss of efficacy of the oncospecific treatment. The drug interactions detected on these 4 patients were at the level of oral absorption of the oncospecific treatment, with the subsequent risk of reduction of its efficacy.

The recommendation for patients with clinical indication for PPIs who must initiate oncospecific treatment with gefitinib, lapatinib or dasatinib is to replace the PPI by an antacid drug administered at least two hours before or two hours after taking these drugs¹². In the case of erlotinib, the PPI can be replaced by an antacid, administered two hours before or two hours after taking the drug, or by an anti-ulcer drug, such as an anti-H₂ drug (H₂ receptors antagonist), administered 12 hours before or after taking the drug¹².

The previously described recommendations were followed for the 4 patients with possible interactions between the PPI and the oncospecific treatment. In one of the patients, who had to initiate treatment with erlotinib, the interaction was solved through discontinuation of the PPI, because its use was not indicated. The two remaining patients who had to initiate treatment with erlotinib presented indication for the use of PPIs, one for gastric protection and the other for non-ulcerous dyspepsia. In both cases, the interaction was solved replacing the PPI by an anti-H₂ anti-ulcer drug, administered with a 12-hour interval of taking erlotinib¹². Finally, in the patient with a medical history of peptic ulcer who required treatment with dasatinib, it was decided in consensus with physicians, to replace the PPI by an antacid, because anti-H₂ drugs will also interact with the oral absorption of dasatinib. In this case, it was more relevant to ensure the efficacy of dasatinib than gastric protection. In those patients for whom the interaction recommends the discontinuation of the PPI, and there is no equally effective alternative without interaction, the decision must be made based on the risk-benefit of PPI discontinuation.

In conclusion, the prescription of a PPI when it is not indicated will lead to drug interactions and potential adverse effects derived of its prolonged use.

It is important the pharmacist to detect unappropriated prescriptions of PPIs, within the overall setting of a therapeutic simplification strategy, through clinical interviews with patients conducted in the Pharmacy Unit, a complete review of medical history, and consultation with the specialist or primary care physician responsible for the patient.

No PPIs should be prescribed without a clinical indication justifying treatment, and even when this is present, the risk-benefit of treatment should always be assessed. For this reason, it is important the pharmacist to promote PPI deprescription in asymptomatic patients without a clear indication for treatment.

 

Bibliography

1. Sánchez M, Fernández E, Alcacera MA, Matarredona A, Mozota J, Guarc E. Evaluación del grado de cumplimiento de un protocolo sobre utilización racional de inhibidores de la bomba de protones en el medio hospitalario. En: 58 Congreso de la Sociedad Española de Farmacia Hospitalaria. 22-25 octubre 2013, Málaga. Farm Hosp. 2013;37(Supl 1):328-9.         [ Links ]

2. Raghunath AS, O'Morain C, McLoughlin RC. Review article: the long-term use of proton-pump inhibitors. Aliment Pharmacol Ther. 2005;22(Suppl 1):55-63.         [ Links ]

3. Van Soest EM, Siersema PD, Dieleman JP, Sturkenboom MCJM, Kuipers EJ. Persistence and adherence to proton pump inhibitors in daily clinical practice. Aliment Pharmacol Ther. 2006;24(2):377-85.         [ Links ]

4. García del Pozo J. Estudio de utilización de antiulcerosos en España (2000-2008). Inf Ter Sis Nac Salud. 2009;33:49-54.         [ Links ]

5. National Comprehensive Cancer Network. Clinical Practice Guidelines in Oncology. Antiemesis. Versión 2. 2014. Disponible en: http://www.nccn.org/professionals/physician_gls/pdf/antiemesis.pdf (último acceso 13 noviembre 2014).         [ Links ]

6. Riesgo de hipomagnesemia asociado a los medicamentos inhibidores de la bomba de protones. Nota informativa de la Agencia Española de Medicamentos y Productos Sanitarios. Fecha de publicación: 23 de diciembre de 2011. Disponible en: http://www.aemps.gob.es/informa/notasInformativas/medicamentosUsoHumano/seguridad/2011/docs/NI-MUH_27-2011.pdf (último acceso 13 noviembre 2014).         [ Links ]

7. Cundy T, McKay JD. Proton pump inhibitors and severe hypomagnesemia. J Current Opinion in Gastroenerology 2011;27:180-85.         [ Links ]

8. Inhibidores de la bomba de protones (IBP): riesgo de fracturas óseas. Informe mensual sobre medicamentos de uso humano y productos sanitarios. Agencia Española de Medicamentos y Productos Sanitarios. Última actualización: 10 de marzo de 2012. Disponible en: http://www.aemps.gob.es/informa/boletinMensual/2012/marzo/informe-medicamentos.htm (último acceso 13 noviembre 2014).         [ Links ]

9. Eom CS, Jeon CY, Lim JW, Cho EG, Park SM, Lee KS. Use of acid-suppressive drugs and risk of pneumonia: a systematic review and meta-analysis. CMAJ. 2011 Feb 22;183(3):310-9.         [ Links ]

10. Linsky A, Gupta K, Lawler EV, Fonda JR, Hermos JA. Proton pump inhibitors and risk for recurrent Clostridium difficile infection. Arch Intern Med. 2010:170(9):772-8.         [ Links ]

11. Shahab I, Patterson WP. Renal and Electrolyte Abnormalities Due to Chemotherapy. In: Perry MC, ed. The Chemotherapy Source Book. 4^th ed. Philadelphia: Lippincott Williams & Wilkins; 2008:223-33.         [ Links ]

12. Guía de Interacciones Farmacológicas de Citostáticos. Grupo de Farmacia Oncológica de la Sociedad Española de Farmacia Hospitalaria. Última revisión Febrero 2014. Disponible en: http://interacitoral.sefh.es (último acceso 13 noviembre 2014).         [ Links ]

 

 

Dirección para correspondencia:
Correo electrónico: mpujal@cst.cat
(Meritxell Pujal Herranz).

Recibido el 25 de agosto de 2015;
aceptado el 9 de mayo de 2016.