 |
 |
 |
 |
 |
 |
 |
|
|
|
|||||||||
|
|||||||||||
Clinical Assistant Professor, Faculty of Pharmacy, University of Montreal, Montreal, Quebec, Canada; Pharmacist, Internal Medicine/Gastroenterology, McGill University Health Centre—Montreal General Hospital, Montreal
Clinical Pharmacist, Adult Internal Medicine/Infectious Diseases; Clinical Assistant Professor, University of Michigan Hospitals/College of Pharmacy, Ann Arbor, MI
at time of writing, PharmD Student, College of Pharmacy, University of Michigan
at time of writing, PharmD Student, College of Pharmacy, University of Michigan
Reprints: Dr. Regal, Department of Pharmacy Services, University of Michigan Hospital, UH/B2D301 Box 0008, 1500 E. Medical Center Dr., Ann Arbor, MI 48109-0008, fax 734/936-7027, reregal{at}umich.edu
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
OBJECTIVE: To assess the indications and prevalence of acid suppressants used by inpatients on admission and at discharge.
METHODS: A retrospective chart review of 213 patients admitted to the University of Michigan Hospital non-critical care general medical service was conducted. Relevant medical history, acid suppressant drug used, and indications were collected from both inpatient medical records and discharge medication lists.
RESULTS: Of the 213 patients reviewed, 29% were taking acid suppressants prior to admission, with 33% being proton pump inhibitors (PPIs). Once patients were admitted, acid suppressant use increased to 71% (152 of 213), with 84% PPIs, 11% histamine2-receptor antagonists, and 5% combination therapy. Based upon our criteria, only 10% (15 of 152) of those on acid suppressants were found to have an acceptable indication. In patients where any history of gastroesophageal reflux disorder (GERD) was deemed as an acceptable indication (32 other patients), 31% (47 of 152) had an acceptable indication. For the 137 patients with non-accepted indications, 29% had no discernable indication and 38% were prescribed acid suppressants for corticosteroid-associated or stress ulcer prophylaxis. A history of gastrointestinal bleeds or peptic ulcer disease of more than 3 months since initial diagnosis or documented exacerbation of symptoms comprised 8% of the population. The aforementioned group of GERD patients made up 23% of this group. Compared to the 29% of patients taking acid suppressants prior to admission, 54% (115 of 213) of patients were prescribed acid suppressants at discharge. If only recent exacerbations of GERD were deemed as long-term indications, 10% (12 of 115) of these patients were found to have accepted indications. If all GERDs were acceptable long-term indications, 27% (31 of 115) would have met criteria for acceptable outpatient use.
CONCLUSIONS: There is considerable excess usage of acid suppressants in both the inpatient and outpatient settings.
Key Words: H2-receptor antagonists, proton pump inhibitors, ulcer prophylaxis
Published Online, June 27, 2006. www.theannals.com, DOI 10.1345/aph.1G703
The first PPI, omeprazole, emerged on the US market in the late 1980s. Since then, PPIs have demonstrated superior efficacy in acid suppression over their progenitor agents, the H2RAs. In addition, the PPIs have been found to elicit a relative lack of serious adverse effects or drug interactions. PPIs induce their reduction of gastric acid secretion via the selective and irreversible inhibition of H/K-adenosine triphosphatase, an enzyme within the gastric parietal cells with a half-life of about 50 hours.1 Compared with the PPIs, H2RAs possess some pharmacologic shortcomings. For instance, H2RAs inadequately suppress gastric acid from the parietal cell and only block the histamine2-related stimulation of acid secretion, thereby resulting in tachyphylaxis beginning as early as 48-72 hours into therapy.2 In general, however, both PPIs and H2RAs remain options for therapy in a myriad of acid-peptic conditions.
Both the burgeoning use of PPIs and a small but possibly significant link to certain types of infections such as Clostridium difficile colitis3,4 and community-acquired pneumonia5 have brought forth new reasons to scrutinize the ways in which acid suppressants are prescribed.6 These studies also linked the H2RAs to increased rates of infection. Pneumonia was once thought to be potentiated by H2RA usage in the critically ill for stress ulcer prophylaxis. Although H2RAs are no longer implicated in the development of pneumonia in this arena, the increased usage of PPIs for this role may provide new causes for concern.
It is important to recognize that with the augmentation in the role of acid suppressants, fostered in part by the increased availability of PPI formulations and expanding over-the-counter status, appropriate use of acid suppressant agents requires more scrutiny than ever before. Moreover, the issue with acid suppressants of inappropriate prescribing and polypharmacy remains important for the laudable global mission of optimizing pharmaceutical care and reducing medication costs.
The aim of this study was to assess the indications and prevalence for use of acid suppressants (PPIs and H2RAs) in non-critically ill general medicine patients at the University of Michigan Hospital, both upon admission and at discharge.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
At the time of the review, the hospital formulary included pantoprazole and ranitidine as the preferred PPI and H2RA, respectively. Patient data upon admission included necessary demographics, past medical history, and medication usage. The initiation or addition of acid suppressants was recorded along with their indications, and all changes were followed until hospital discharge. Discharge medication lists were used to identify subsequent outpatient prescribing. If an acid suppressant was listed as an outpatient discharge medication, it was assumed that the patient took that drug following discharge for at least that one prescription fill, but no other records were accessed to verify or refute outpatient use of these drugs.
INDICATIONS FOR ACID SUPPRESSANT THERAPY
Accepted indications were those currently listed by the Food and Drug
Administration (FDA) or those that have support of strong evidence in the
medical literature. The following were considered accepted indications for
acid suppressant therapy: gastric and duodenal ulcer with documented
exacerbations within the last 3 months, GERD with documented exacerbations
within the last 3 months, pathological hypersecretory conditions, symptoms
recently associated with indigestion within the last 3 months, and PPI use for
Helicobactor pylori eradication; prophylaxis for gastropathies
associated with nonsteroidal antiinflammatory drugs (NSAIDs) were also deemed
appropriate.
Since most studies looking at acid suppressant treatment for healing of
esophageal and peptic ulcers lasted about 4-8 weeks, we arbitrarily chose
continuation of treatment of these disorders beyond 3 months (
12 wk),
without documentation of disease exacerbation, to be inappropriate. Since it
was also noted that a fairly large subgroup of patients with GERD may require
extended and intermittent therapies with PPIs beyond the initial 1-3 months of
acute therapy, we analyzed the data as if all patients with ongoing diagnoses
of GERD may have possessed a suitable long-term indication.
Non-accepted indications included prescribing of acid suppressant medications for low-risk stress ulcer prophylaxis for non-critically ill medical patients; prophylaxis of PUD associated with corticosteroids or anticoagulants; a history of GERD, gastrointestinal (GI) bleed, or PUD for more than 3 months without ongoing complications or exacerbations; anemia; and no apparent indication discernable from the patient chart.
As previously stated, patients with a history of GERD but no documented exacerbations within the 3 months prior to admission were included in 2 separate analyses. One version assumed that only recent GERD exacerbations were appropriate indications for acid suppressant use, and the other assumed that any patient with even a remote history of GERD had an acceptable justification for use.
STATISTICS
Descriptive statistics were used to describe observations of comparisons.
SPSS software was used to process the data.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
|
|
As shown in Table 3, 71% (n = 152) of patients included in the analysis were taking acid suppressant drugs while in the hospital, with PPIs used in most patients. If indications for patients with a listed history of GERD but no exacerbations documented within the last 3 months were deemed unacceptable, acceptable indications for acid suppressant therapy accounted for 10.5% of the group. If all GERD patients were deemed to have acceptable indications, the number increased to 30.9%. Therefore, depending on the way in which GERD was considered in the analysis, 69.1-89.5% of acid suppressant usage was for unacceptable indications.
|
Of the 137 unacceptable indications for use of acid suppressants, 29 of the medical records indicated that the drugs were prescribed for prophylaxis of corticosteroid-related ulcer and 23 for prophylaxis of other reasons for stress ulcer, with these 2 groups totaling 38% of all nonindicated use (Table 4). A previous history of PUD, GERD, or GI bleed, but without current symptoms or reason for admission within the last 3 months, accounted for 31% of inappropriate acid suppressant usage.
|
Table 3 also presents the findings for the sample of inpatients upon their discharge. Inpatients who continued taking acid suppressants at discharge included 54% of the total population. However, if all patients with even a remote history of GERD were included as having an acceptable indication, then 27% (n = 31) of patients would have had outpatient therapy deemed acceptable. Thus, the vast majority of discharged patients were prescribed acid suppressants without acceptable indications. The distribution of outpatient acid suppressants was mostly PPIs, followed by H2RAs and combination PPI-H2RAs.
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
Indeed, based on our findings and those of previous studies,7,15-20 prescribing of acid suppressant medications appears to be relatively ubiquitous and inappropriate. A 7 month retrospective review similar to ours found that only 37% of the medical inpatient population from a major Australian teaching hospital was receiving PPIs for indications deemed acceptable by the Australian Schedule of Pharmaceutical Benefits.15 A US study observed that 54% of general medicine inpatients at a teaching institution were prescribed acid suppressants. Of those patients, 65% did not have an accepted indication for therapy, as per the FDA or strongly supported medical evidence. Our figures may therefore reflect an increasing trend in inappropriate prescribing of acid suppressants since 2000,15,16 particularly PPIs. This trend was reflected in a study by Bashford et al.21 in the UK, observing that PPI prescribing for indications unlicensed by the UK Medicines and Healthcare products Regulatory Agency increased in the 1990s and consisted of 46% of new PPI prescriptions in 1995.
An examination of the non-accepted indications for acid suppressants provides some insight into clinician perspectives on acid suppressants. There appears to be an unfounded concern for low-risk inpatients developing stress ulcers. Nardino et al.16 also recognized this, as 41% (33 of 80) of their non-intensive care unit medical patients without a true indication were taking acid suppressants for stress ulcer prophylaxis. Our report demonstrated a rate of 38% of all inappropriate prescribing being linked with either corticosteroid use (21%) or other indications for stress ulcer prophylaxis (17%). Looking at it another way, 34% of all acid suppressant use in our study was justified based on these 2 unaccepted indications. It is also quite likely that a number of the cases in which no indication was documented (29% of the non-indicated uses) could have been prescribed an acid suppressant for prophylaxis, thus having this stress ulcer prophylaxis group encompass half or more of all acid suppressant use during admission.
Data are lacking to support general PUD prophylaxis for medical inpatients outside of the critical care environment. In contrast with the evidence supporting PPI protection from PUD for those on long-term NSAIDs,22 there is also a relative paucity of evidence for the prophylaxis of PUD for those on corticosteroids. A meta-analysis found no increase in the risk of PUD associated with corticosteroid use.17 Additionally, a more recent survey in an Italian teaching hospital observed that 40% of inappropriate prescribing of PPIs involved ulcer prophylaxis in low-risk patients.18 The use of acid suppressants for prophylaxis of PUD for low-risk medical patients should be questioned. Conversely, the PPIs' role in acutely treating PUD is evident. However, the benefit of PPIs for the long-term maintenance of PUD is not so clear.
A few other studies have also supported our findings of inappropriate use of acid suppressants. For instance, Walker et al.19 found that 67% of PPIs were prescribed for unapproved indications in their study involving hospital inpatients. In addition, a prospective interventional analysis documented that 66% of active prescriptions for lansoprazole twice daily from medical records did not comply with the medical center's guidelines.20
Another salient observation in our study was the continued misuse of acid suppressants upon hospital discharge. The findings indicated that these agents were being used in only 29% of the study population before admission; however, 54% of the patients were prescribed them upon discharge, with only a small percentage for accepted indications. Of patients discharged on acid suppressive therapy with no acceptable indication but with a history of GERD acceptable indication, 79.1% used PPIs, 14.8% used H2RAs, and 6.1% used combination PPI-H2RAs. These findings once again mirror those of Nardino et al.,16 whereby 55% of their patients were discharged on acid suppressants, with the majority of those not having an appropriate indication.
In our study, outpatient prescriptions seemed to mimic the inpatient trends, with our determined range of appropriateness anywhere from 10% to 27%. Thus, inpatient misuse seemed to enable continued misuse in the outpatient environment. Assiduous discharge medication reconciliation by pharmacists and other members of the multidisciplinary care team should be encouraged to curb this tendency.
Given their significant cost to the currently cost-overloaded healthcare system and the recent data associating PPIs with increased risk of C. difficile colitis and community-acquired pneumonia,3-6 the prescribing of PPIs in both the inpatient and outpatient settings needs to be more critically questioned using the time-honored benefits versus risks adage. Certainly, hospital formulary and outpatient care guidelines should be formulated to help implement more prudent and sparing use of acid suppressant therapy for both inpatients and outpatients. When a PPI or any other acid suppressant agent is included on a medication list, the pharmacist should carefully review the patient's medical history. Finally, prescribers need to be challenged to stop the drug if the history cannot support continued use, and a drug holiday should at least be attempted in equivocal situations.
|
Limitations |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
Our inability to definitively include or exclude patients with GERD who were using acid suppressants for longer than 3 months demonstrates our limitation to analyze those on long-term, intermittent, on-demand, or refractory regimens. However, there is no consensus on what constitutes long-term acid suppressant prescribing. Our definitions have taken into account FDA approvals and strong evidence in this area. Our data reflect that 26% of our accepted indications were for patients with GERD who were using acid suppressants less than 3 months after the last documented episode. Conversely, 23% of our unaccepted indications were for patients with GERD using acid suppressants longer than 3 months but without documentation of active symptomatic disease. Since a large percentage of patients with GERD who were treated with an initial several week course experienced relapses and required intermittent or continuous therapy to control the disease, the number of patients studied who had truly appropriate acid suppressant prescribing was likely somewhere between 15 and 47. Thus, we had fairly wide ranges of appropriateness: 10-31% in the hospital setting and 10-27% in the outpatient setting.
|
Conclusions |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  I. Forgacs and A. Loganayagam Overprescribing proton pump inhibitors BMJ, January 5, 2008; 336(7634): 2 - 3. [Full Text] [PDF]
|
|
|
|
 |
|
 R. R. A. Grube and D. B. May Stress ulcer prophylaxis in hospitalized patients not in intensive care units Am. J. Health Syst. Pharm., July 1, 2007; 64(13): 1396 - 1400. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
 |
 |
 |
 |
