Each year we review therapeutic issues covered in the past and identify new clinical trials that reinforce or change our recommendations. Our last Review and Update, Letter 9, was in September 1995. In this issue we review thirteen recent clinical trials that pertain to nine previous Letters.
Many antibiotic regimens have been tested for the eradication of H. pylori. Thus far none have achieved superior eradication rates to our original recommendation, the one week course of triple therapy (A). Other one week regimens which have achieved >90% eradication rates are shown as well (B).
30 ml QID
250 mg QID
|+||tetracycline 500 mg
amoxicillin 500 mg QID
|lansoprazole 30 mg
omeprazole 20 mg BID
500 mg BID
|+||amoxicillin 1 g BID
metronidazole 400 mg BID
The new proton pump inhibitor lansoprazole has been shown in a randomised controlled trial (RCT) to be similarly effective in doses of 15 mg or 30 mg daily for the maintenance treatment of reflux esophagitis.(1)
The potential risks of long-term proton pump inhibitor therapy was demonstrated in a cohort study (2) comparing patients with reflux esophagitis treated with fundoplication (a surgical technique to prevent reflux) or chronic omeprazole, 20-40 mg daily. Patients were followed for an average of 5 years; severity of histologic gastritis worsened overall and atrophic gastritis developed in 30% of patients in the omeprazole group who were H. pylori positive at baseline. Atrophic gastritis did not develop in any of the patients treated with fundoplication whether or not they were infected with H. pylori at baseline (p<0.001). Long-term acid suppression with proton pump inhibitors is associated with worsened H. pylori induced inflammatory gastritis, leading to atrophic gastritis and possibly to increased risk of dysplasia and gastric cancer.
In a large cohort study (3) of peptic ulcer patients, followed for an average of 9 years, the risk ratio for gastric cancer among patients with gastric ulcer and duodenal ulcer was 1.8 (1.6 to 2.0) and 0.6 (0.4 to 0.7), respectively. This creates a seeming paradox since both gastric and duodenal ulcers are associated with H. pylori infection. One explanation (see Editorial (3)) is that infection with H. pylori in childhood leads over time to atrophic gastritis, low acid production, and increased risk of gastric ulcer and gastric cancer in adulthood, whereas infection with H. pylori as an adult leads to increased acid secretion, an increased risk of duodenal ulcer, and initially a decreased risk of gastric cancer.
Seven recent studies pertaining to calcium channel blockers (CCBs) add to the previous suggestive evidence that CCBs increase the risk of myocardial infarction as compared to beta blockers (Letter 9). Five of these studies demonstrated an increased risk of CCBs as compared to other antihypertensives or placebo, one study showed no difference as compared to placebo, and one study showed a benefit as compared to placebo. The magnitude of the risks and benefits are shown in the Table. These studies suggest that there are significant additional risks associated with CCBs: mortality, hemorrhage, and cancer, but they do not distinguish between the different CCBs. The designs of the 7 trials are listed below.
The isradipine vs hydrochlorothiazide (MIDAS) trial is a randomised double-blind trial of 883 patients (average age 58.5 yr.) with mild hypertension.(4) The primary outcome measure, ultrasound evidence of atherosclerosis in carotid arteries was not different (p=0.68) in the two treatment groups. There was a higher incidence of vascular events and procedures in the patients receiving isradipine, an intermediate acting CCB (see Table).
The nimodipine trial was a randomised double-blind, placebo controlled trial designed to determine whether nimodipine (a short acting CCB) administered perioperatively reduces the incidence of new neurological deficits associated with cardiac valve replacement.(5) New deficits were observed in 77% of the nimodipine group (p=0.55) versus 72% of the placebo group. Because of increased mortality in the nimodipine patients (see Table) the study was terminated early. Five of the 10 deaths in the nimodipine group were associated with major bleeding. All CCBs interfere with blood coagulation by inhibiting platelet aggregation.
The three prospective cohort studies(6),(7),(8) demonstrate the potential risks of CCBs (including short acting verapamil, diltiazem and nifedipine) in an elderly (>70 yr) North American population between 1985 and 1992. These cohorts were followed for different periods to quantitate the different outcome measures shown in the Table. The postulated mechanism whereby CCBs increase risk of cancer is by inhibition of apoptotis, a natural method by which cells sacrifice themselves to prevent cancer. The same outcomes in patients taking ACE inhibitors (not shown) were not significantly different from those in patients taking beta blockers.
The amlodipine trial (PRAISE) compared amlodipine (a long acting CCB) with placebo in patients with severe CHF already treated with digoxin, diuretics, and ACE inhibitors.(9) Amlodipine caused a statistically significant 2 mmHg drop in blood pressure, however there was no statistically significant difference in the primary endpoints (see Table). In a subgroup analysis all the benefit occurred in the patients with nonischemic dilated cardiomyopathy (RR 0.69) with no effect in patients with ischemic heart disease (RR 1.04). As a result a follow-up trial has been initiated in patients with nonischemic dilated cardiomyopathy (PRAISE-2).
The Shanghai trial of nifedipine (10-30 mg BID) in the elderly (STONE) has a number of flaws in design and process that could lead to bias: single-blind, alternate (not random) allocation, reallocation of 74 patients from placebo to nifedipine group, unequal distribution of men and women (p<0.01) and many patients lost to follow-up (16.5% in the nifedipine group and 21.4% in the placebo group at 24 months). The results in the Table are based on the incidence of terminating events in the group of patients actually receiving nifedipine or placebo therapy. There was no significant difference in MI, total deaths or in cardiovascular deaths, between the two groups.
|Clinical Trial||Trial||Event incidence %||RR||RRR||ARI/R||NNH/T||Trial Duration|
|(measured outcome events)||Design||Drug/Placebo||CCBs||%||%||in years|
|Isradipine vs HCTZ in Hypertension (4)
(any vascular event or procedure)
|Perioperative Nimodipine vs Placebo (5)
|CCBs vs beta blockers in hypertension
|CCBs vs beta blockers in hypertension
|CCBs vs beta blockers in hypertension
|Amlodipine vs. placebo in severe CHF
(total mortality or cardiovascular morbidity)
|Nifedipine vs. placebo in elderly
hypertensives (10) (total terminating events)
RR=relative risk || RRR=relative risk reduction || ARI/R=absolute risk increase/reduction || NNH/T=number needed to harm/treat || CCT=clinical controlled trial || RCT=randomised control trial || *=not significant
The weight of evidence at the present time suggests that CCBs (both short and long acting varieties) at best provide less benefit than other available antihypertensive drugs, including thiazides, beta blockers and ACE inhibitors, and that they may be conferring unnecessary risks to our patients. It is therefore important to reevaluate patients taking CCBs and continue them only if there is no satisfactory alternative therapy.
Information about the potential risks of CCBs was first presented in the Spring of 1995. Since that time clinicians and the public have been exposed to large amounts of information surrounding the controversy. Up to and including 1994, there was a steady increase in the number of prescriptions for CCBs. In 1995 this increase plateaued, and in 1996 (Jan.-Jun.) there was a 14% decrease in the number of patients receiving CCBs as compared to 1994. If one identifies new patients, 66 years and over prescribed one of the 4 major classes of antihypertensives, there has been a 40% decrease of CCB prescriptions in 1996 as compared to 1994. However, CCBs still account for 14.5% of first line therapy to new patients 66 years and over, and they remain second only to the diuretics in total number of prescriptions of cardiovascular drugs in 1996.
In the Asthma letter we reported that the administration of regularly scheduled fenoterol is associated with worsening of asthma control as compared to when it is taken on an as needed basis. In North America fenoterol is not commonly used and salbutamol (albuterol) is the most common beta-2-agonist. In a recent RCT, regularly scheduled salbutamol was compared with as needed use for mild asthma.(11) Over a 16 week treatment period measures of asthma control were not significantly different between the two schedules; it is clear that beta-2-agonists should be prescribed on an as needed basis for most patients.
There were two cohort studies published shortly after our Hormone Therapy Letter. The first is a further analysis of the Nurses Health cohort study, extending the follow-up to 16 years.(12) The only new data from this study is that the patients who took estrogen plus progestin had if anything a lower risk of coronary heart disease than women taking estrogen alone. Hormone therapy was associated with a non-statistically significant increased risk of stroke. When the benefits and risks were related to dose, the patients on the lower doses of conjugated estrogens (0.3 and 0.625 mg) consistently did better than those on higher doses. The second cohort study measured the risk of Alzheimer's Disease in 1124 elderly patients in New York city.(13) In the 12.5% of these patients who reported taking estrogen after menopause there was a significantly reduced risk 0.4 (0.22-0.85) of Alzheimer's disease. Data from cohort studies is useful for generating testable hypotheses, but should not be used as justification for prescribing hormone therapy. Fortunately, randomised controlled trials have been initiated to test these hypotheses.
Robinson M, Lanza F, Avner D, Haber M. Effective treatment of reflux esophagitis with low- dose lansoprazole. A randomized, double-blind, placebo-controlled trial. Ann Intern Med. 1996; 124:859-867.
Kuipers EJ, Lundell L, Klinkenberg-Knol EC, et al. Atrophic gastritis and Helicobacter pylori infection in patients with reflux esophagitis treated with omeprazole or fundoplication. N Engl J Med. 1996; 334:1018-2.
Hansson LE, Nyren O, Hsing AW, et al. The risk of stomach cancer in patients with gastric or duodenal ulcer disease. N Engl J Med. 1996; 335:242-9, and Editorial 278-280.
Borhani NO, Mercuri M, Borhani PA, et al. Final outcome results of the Multicentre Isradipine Diuretic Atherosclerosis Study (MIDAS). JAMA. 1996; 276:785-791.
Legault C, Furberg C, Wagenknecht LE, et al. Nimodipine neuroprotection in cardiac valve replacement Report of an early terminated trial. Stroke.1996; 27:593-598.
Pahor M, Guralnik JM, Corti MC, et al. Long-term survival and use of antihypertensive medications in older persons. J Am Geriatr Soc. 1995; 43:1191-1197.
Pahor M, Guralnik JM, Furberg CD, et al. Risk of gastrointestinal haemorrhage with calcium antagonists in hypertensive persons over 67 years old. Lancet. 1996; 347:1061-65.
Pahor M, Guralnik JM, Salive ME, et al. Do calcium channel blockers increase the risk of cancer? Am J Hypertens. 1996; 9:695-699.
Packer M, O'Connor CM, Ghali JK, et al. Effect of amlodipine on morbidity and mortality in severe chronic heart failure. N Engl J Med. 1996; 335:1107-14.
Lansheng Gong, Weizhong Zhang, Yijun Zhu, et al. Shanghai trial of nifedepine in the elderly (STONE). Journal of Hypertension. 1996; 14:1237-1245.
Drazen JM, Israel E, Boushey HA, Chinchilli VM, et al. Comparison of regularly scheduled with as-needed use of albuterol in mild asthma. N Engl J Med. 1996; 335:841-7.
Grodstein F, Stampfer MJ, Manson JE, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N Engl J Med. 1996; 335:453-61.
Please send feedback!
Back to Therapeutics Letter.
Back to the Therapeutics Initiative Home Page.
©Therapeutics Initiative. Last updated: December 6, 1996.