Inappropriate Physician Habits in Prescribing Oral Nifedipine Capsules in Hospitalized Patients

Transcription

1 A]H 1996;9: Inappropriate Physician Habits in Prescribing Oral Nifedipine Capsules in Hospitalized Patients Faiz Rehman, George A. Mansoor, and William B. White Despite the absence of an approved Food and Drug Administration (FDA) indication, the use of oral or sublingual nifedipine for hypertension in the hospitalized patient has become an increasingly common practice. The purpose of the study was to assess the clinical circumstances for which the drug was being prescribed and the practices of attending and resident physicians. Dosing of oral nifedipine capsules in medical and surgical inpatients was studied prospectively in three central Connecticut hospitals (University, community-teaching, and private nonteaching) during a 60-day period from January to March, Through evaluation of computerized pharmacy and medical records, data were collected on diagnostic reasons for ordering nifedipine, pre- and posttreatment blood pressures, dosing frequency, clinical documentation associated with drug prescription, and adverse events. Physicians and nurses at the respective hospitals were unaware of the conduct of the study. The incidence of nifedipine capsule administration at all three hospitals was 3.4% (152 dosings in 83 patients14489 hospitalized patients/ 60 days). Practice habits and blood pressure changes differed minimally among hospitals and physicians. Sixty-three percent of nifedipine orders were given over the telephone for arbitrary and asymptomatic blood pressure elevations and 98% of the orders lacked bedside patient evaluation. Follow-up of the blood pressure was performed within 3 h in 51% of patients and within 2 h in 24%, while in 25% there was no documentation of follow-up until 2 to 6 h after nifedipine dosing. Mean pretreatment blood pressure was 186/ mm Hg (range: 150 to 260 mm Hg systolic and 50 to 125 mm Hg diastolic). Blood pressure fell 32/ /16 mm Hg (range: -92 to +8 mm Hg systolic and -90 to +28 mm Hg diastolic) and was related to the level of pretreatment blood pressure tr = 0.53 for systolic blood pressure, and r = 0.49 for diastolic blood pressure, P <.OOl for both). Large, asymptomatic blood pressure reductions were common. One hypertensive patient experienced severe hypotension accompanied by an anterior wall myocardial infarction. These data demonstrate inappropriate physician prescribing of oral nifedipine in hospitalized patients characterized by a lack of proper assessment prior to drug dosing, highly arbitrary treatment parameters that were written without regard for symptoms or underlying illnesses, and slow follow-up for evaluation of the clinical response to therapy. Am J Hypertens 1996;9: KEY WORDS: Nifedipine, hypertension, hypertensive emergencies, hypertension urgencies, antihypertensive therapy. Received October 18, Accepted March 25, From the Section of Hypertension and Vascular Diseases, Department of Medicine, University of Connecticut Health Center, Farmington, Connecticut. This work was presented in part at the 10th meeting of the Ameri- can Society of Hypertension, May 18, 1995, New York, NY. Address correspondence and reprints requests to William B. White, MD, Professor of Medicine, Section of Hypertension and Vascular Diseases, University of Connecticut Health Center, Farmington, CT by the American journal of Hypertension, Ltd. Published by Else&r Science, Inc /96/$25.00 PI1 SO (96)00172-O

2 1036 REHMAN ET AL AJH-OCTOBER 1996-VOL. 9, NO. 10, PART 1 N ifedipine, a dihydropyridine calcium channel blocker that lowers systemic vascular resistance, has become a commonly used agent for the treatment of hypertension. In the immediate-release formulation, the rate of absorption of the drug appears to be the greatest after the capsules are bitten through and swallowed rather than swallowed as a whole or bitten through and kept under the tongue.2,3 However, a great deal of variability in pharmacokinetic and pharmacodynamic effects has been observed among subjects following oral dosing studies.3,4 Clinical studies demonstrating the efficacy of nifedipine capsules for hypertensive emergencies or urgencies have been numerous in the medical literature during the past decade.5m8 Despite its common use in acute hypertensive disorders, nifedipine capsules never actually received this indication from the US Food and Drug Administration (FDA). Apparently the FDA was concerned because of the lack of dose-response information and safety of the immediate-release formulation in the severely hypertensive population. Indeed, there have been numerous reports of adverse outcomes following nifedipine capsule dosing, including severe symptomatic hypotension, 0-12 precipitation of coronary ischemia, 3-5 pulmonary edema, I6 and cerebral ischemia.17 Furthermore, excessive lowering of blood pressure by nifedipine may be detrimental in patients with chronic hypertension and underlying cardiac or cerebrovascular disease on other antihypertensive agents. During the past several years, we had noted a progressive and somewhat alarming increase in the use of nifedipine capsules in medical and surgical inpatients at our teaching hospital. It was unknown to us whether this phenomenon was exclusive to physicians-in-training or was true of attending physicians as well. Additionally, the indications and blood pressure levels in patients receiving this oral therapy for hypertension had not been assessed. Thus, we studied the prescribing and practice habits of resident and attending physicians at three area hospitals to evaluate the appropriateness of this therapy in the hospitalized patient. METHODS Hospital Sites This study was conducted in three hospitals in central Connecticut: University of Connecticut Health Center, New Britain General Hospital, and Bristol Hospital. The University of Connecticut Health Center in Farmington, Connecticut is a 232-bed medical school hospital with both internal medicine and surgery residents writing all of the daily orders for hospitalized patients. New Britain General Hospital is a 432-bed community teaching hospital (with occupancy of about 250 beds during the study period) where both resident physicians and private attending physicians write medication orders. The Bristol Hospital is a 250-bed commu- nity nonteaching hospital where all medication orders are written by attending physicians. The patient populations for all of these hospitals are similar and are derived from the metropolitan area of Hartford, Connecticut. This study was approved by the University of Connecticut Institutional Review Board. Data Collection Over a period of 8 weeks (January to March 1994) data were collected prospectively through review of pharmacy and medical records on a daily basis. The study population consisted of all patients on the medical and surgical floors and excluded emergency room patients. The variables evaluated included age, gender, nifedipine capsule dose ordered, route and frequency of administration, indication for ordering nifedipine capsules, method of medication order (written or verbal by telephone), threshold blood pressure values for ordering nifedipine capsules, pre- and posttreatment blood pressures, time (to the nearest 15-min interval) that the pressure was taken after nifedipine dosing, and adverse effects. Physicians, patients, and the nurses at the respective hospitals were not aware of the conduct of the study. Data Analysis The demographic variables, indications for medication orders, threshold blood pressure values for ordering nifedipine, and adverse side effects were tabulated. These variables were compared among hospitals using x2 analysis. The pre- and postdosing systolic and diastolic blood pressures were compared using paired t testing and changes between the hospitals were compared by analysis of variance. Relationship between the pretreatment blood pressure and change in blood pressure postnifedipine dosing was assessed by Pearson s correlation testing. RESULTS One hundred and fifty-two doses of nifedipine were administered to 83 hospitalized patients during the study period. The mean age of the patients was 70 years (range 28 to 100 years) and there were 40 men and 43 women. The incidence of nifedipine capsule administration for all three hospitals was 3.4% (152 doses in 83 patients / 4489 patients discharged). Practice habits, blood pressure thresholds for ordering the nifedipine capsule, and blood pressure changes postdosing did not differ significantly among the three different hospitals with the exception that the age was lower and the pretreatment diastolic pressure higher at the university hospital (Table 1). The most commonly prescribed dose was 10 mg (96%). However, multiple 10 mg doses were administered to 65% of the patients. Additionally, 63% of the nifedipine capsule orders were given verbally over the phone for arbitrary and asymptomatic blood pressure

3 AJH-OCTOBER 1996-VOL. 9, NO. 10, PART 1 INAPPROPRIATE PRESCRIBING PRACTICES 1037 TABLE 1. DEMOGRAPHIC AND BLOOD PRESSURE DATA AMONG THE THREE STUDY HOSPITALS University of Connecticut Health New Britain General Parameter Center Hospital Bristol Hospital All Hospital type University-Teaching Community-Teaching Private-Nonteaching Prescribing physicians Medical and surgical Medical and surgical house Private attendings house staff staff and attendings Medical/surgical discharges during study period Nifedipine dosingst 22 (4.2%) 96 (3.1%) 34 (3.7%) 152 (3.4%) (number, incidence) Number of patients receiving 16 (3.07c) 45 (1.5%) 22 (2.1%) 83 (1.8%) nifedipine (number, %) Age, years L r 17 Pretreatment blood pressure Systolic BP, mm Hg ? 20 Diastolic BP, mm Hg Mean change in blood pressure Systolic BP, mm Hg * Diastolic BP, mm Hg i t P <.05 v other 2 hospitals. t More than 1 dose was given to certain pafienfs during the hospitalization elevations and 98% of the orders lacked any evidence that a bedside evaluation had been performed. Thus, it was not possible to evaluate the clinical indications for ordering nifedipine capsules other than blood pressure levels. The distributions of the pretreatment systolic and diastolic blood pressures are shown in Figure 1. Mean pretreatment blood pressure in the entire group was 186/94 i- 20/ 16 mm Hg (Table 1). The threshold systolic blood pressure for ordering nifedipine capsules in hospitalized patients ranged from as low as 150 to as high as 260 mm Hg, while for diastolic pressure the range was 50 to 125 mm Hg. In fact, nearly one-third of the administrations of nifedipine capsules were in patients with a normal diastolic blood pressure. Often times, especially in the records of surgical inpatients, standing, as needed orders were written by physicians for nurses to carry through based on arbitrary blood pressure values (eg, 10 to 20 mg nifedipine orally every 4 h for a systolic blood pressure >160 mm Hg). The follow-up of the blood pressure response to nifedipine dosing by a clinician was performed within 1 h in 51% of the patients and within 2 h in 24% of the patients, while in 25% there was no documentation of follow-up until 2 to 6 h after nifedipine dosing. Blood pressure fell -32 / / 16 mm Hg (range -92/ -90 to +8/ +28 mm Hg) and was moderately related to the level of pretreatment blood pressure (r = 0.53, P <.OOOl for systolic blood pressure, and r = 0.49, P <.OOl for diastolic blood pressure). There was also a modest but significant relationship between age and reduction in systolic blood pressure (r = 0.17, P =.04). One serious adverse event occurred in an asymptomatic hypertensive patient with history of angina pectoris: severe hypotension accompanied by anterior wall myocardial infarction was documented approximately 1 h following dosing of 20 mg of nifedipine. DISCUSSION This study is the first to examine the prescribing habits of physicians regarding nifedipine capsules and hospitalized patients with elevations of blood pressure. Unfortunately, it appears that the decision to administer nifedipine capsules was based on arbitrary and wide ranges of blood pressures set by individual physicians (Figure 1). Furthermore, most of the nifedipine orders (63%) were given verbally over the phone and 98% of the medication orders lacked clinical bedside evaluation. Although most patients with truly hypertensive emergencies characterized by acute target organ injury will have diastolic blood pressures exceeding 120 mm Hg, the converse is not generally true.l Evidence of new or progressive end organ damage, rather than an absolute level of blood pressure, should qualify the need for urgent control or acute reduction. Hence in these situations performance of a complete physical examination, evaluation of the patient s symptoms and review of prior medical records is imperative.l Both animal and human studies have shown upper and lower limits of blood pressure for cerebral blood flow autoregulation but these end points are highly variable depending on the chronicity of hypertension,

4 1038 REHMAN ET AL AJH-OCTOBER 1996VOL. 9, NO. 10, PART 1 whose systolic blood pressures exceeded 200 mm Hg had a 40 to 100 mm Hg drop in pressure. These changes are unacceptably large in many subpopulations of hypertensive patients. CONCLUSIONS Diastolic Blood Pressure, mm Hg FIGURE 1. Threshold blood pressure ualues for oral nifedipine capsule dosing in hospitalized patients. The upper panel shows the distribution of the pretreatment systolic blood pressures and the lower panel shows the distributions of the pretreatment diastolic blood pressures. duration of antihypertensive therapy, and the rapidity of the rise or fall of the blood pressure. The reductions in systolic and diastolic blood pressures observed in our study may not appear to be excessive if compared to prior clinical trials with oral or sublingual nifedipine used for hypertensive emergencies or urgencies. However, it should be recognized that in approximately 50% of the nifedipine administrations, follow-up pressures were measured 2 to 6 h postdosing. As nifedipine s peak pharmacodynamic effect occurs 30 to 90 min postdosing,2-4 our reported changes in blood pressures may have underestimated the actual peak reductions. The correlation between the pretreatment systolic blood pressure and the fall in systolic blood pressure following nifedipine dosing has been previously reported.5,6 Our study also found a correlation between the pretreatment diastolic pressure and change in diastolic blood pressure as well as between age and the change in systolic blood pressure. In fact, patients While nifedipine is likely to be the most effective oral agent for the treatment for severe hypertension, it must also be realized that the indiscriminate use of this drug is poor medical practice that can lead to potentially serious side effects. In a patient deemed to be severely hypertensive, the physician should evaluate the clinical situation comprehensively for the need for rapid-acting antihypertensive drug therapy. For the treatment of severe hypertension, the risk of rapid and excessive reduction of blood pressure with nifedipine may outweigh the treatment benefits. i In most instances of severe, nonaccelerated or symptomatic hypertension, it seems logical to decrease the blood pressure with a long-acting antihypertensive agent and to follow the response closely over several days. The disturbing practices observed in our study, which appear to be widespread among both resident and attending physicians, should be discouraged by pharmacy and therapeutic committees of academic and community hospitals. ACKNOWLEDGEMENTS The authors thank the Pharmacy and Medical Records departments of the University of Connecticut Health Center, New Britain General Hospital, and Bristol Hospital for their cooperation during the conduct of this study REFERENCES Robinson F, Dobbs RJ, Kelsey CR: Effect of nifedipine on resistance, vessels, arteries and veins in man. Br J Clin Pharmacol 1980;10: McAllister RG: Kinetics and dynamics of nifedipine after oral and sublingual doses. Am J Med 1986; 8l(suppl 6A):2-5. Foster TS, Hamann SR, Richards VR, et al: Nifedipine kinetics and bioavailability after a single intravenous and oral doses in normal subjects. J Clin Pharmacol 1983;23: Harten J, Breimer D, Danhoff M, et al: Negligible sublingual absorption of nifedipine. Lancet 1987;ii: Bertel 0, Conen D, Radu E, et al: Nifedipine in hypertensive emergencies: Br Med J 1983;286: Haft J, Litterer WE: Chewing nifedipine to rapidly treat hypertension. Arch Intern Med 1984; 144: McDonald A, Yealy D, Jacobson S: Oral labetelol versus oral nifedipine in hypertensive urgencies in the emergency department. Am J Emerg Med 1993;11: Angeli P, Chiesa M, Caregaro L, et al: Comparison of sublingual captopril and nifedipine in the immediate

5 AJH-OCTOBER 1996VOL. 9, NO. 10, PART I INAPPROPRIATE PRESCRIBING PRACTICES treatment of hypertensive emergencies. Arch Intern 15. Med 1991;151: Messerli F, Kowey P, Grodzzicki T: Sublingual nifedipine for hypertensive emergencies. Lancet 1991;ii: Watchner RM: Symptomatic hypotension induced by nifedipine in the acute treatment of severe hyperten- 17. sion. Arch Intern Med 1987;147: Boden WE, Korr S, Bough EW: Nifedipine induced hypotension and myocardial ischemia in refractory angina pectoris. JAMA 1985;253: Aromatorio G, Uretsky B, Reddy P: Hypotension and sinus arrest with nifedipine in pulmonary hyperten- 20. sion. Chest 1985;87: O Mailia J, Sander G, Giles TD: Nifedipine induced myocardial ischemia or infarction in the treatment of hypertensive urgencies. Ann Intern Med 1987;107: Jariwalla AG, Anderson EG: Production of ischemic cardiac pain by nifedipine. Br Med J 1978;1: Zangerle K, Wolford R: Syncope and conduction disturbances following sublingual nifedipine for hypertension. Ann Emerg Med 1985;10: Gillmer DJ, Kark P: Pulmonary edema precipitated by nifedipine. Br Med J 1980;280: Noble-Orazio E, Sterzi R. Cerebral ischaemia after nifedipine treatment. Br Med J 1981;283:948. Opie LH, White DA: Adverse interaction between nifedipine and beta blockade. Br Med J 1980;281:1462. Ferguson K, Vlasses I?: Hypertensive emergencies and urgencies. JAMA 1986;255: Strangaard S: Autoregulation of cerebral blood flow in hypertensive patients. The modifying influence of prolonged antihypertensive treatment on tolerance to acute drug induced hypotension. Circulation 1976; 53: Fagan T. Acute reduction of blood pressure in asymptomatic patients with severe hypertension: An idea whose time has come-and gone. Arch Intern Med 1989; 149: