Skip navigation



"There is a disorder of the breast marked with a strong and peculiar symptoms considerable for the degree of danger belonging to it and not extremely rare of which I do not recollect any mention among medical authors. The seat of it and sense of strangling and anxiety with which is attended may make it not improperly be called Angina pectoris. Those who are afflicted with it, are seized, while they are walking, and more particularly when they walk soon after eating with a painful and most disagreeable sensation in the breast."

So William Heberden, is some nice 18th century language, first described angina in 1768.

Drawing the line

Bandolier readers who ask about angina don't have problems diagnosing it, but they do ask about how best to treat it. Now there we have to draw the line. It simply is not possible to do that for a multitude of obvious reasons. What Bandolier can do is to look for evidence, signpost it, and précis it. This might not give all the answers: indeed, it never can. But it can help to inform.

For angina we found one large trial which compared a ß-blocker (bisoprolol) with a calcium antagonist (nifedipine).


This refers to the Total Ischaemic Burden Bisoprolol Study [1], which was a multicentre trial comparing bisoprolol and nifedipine. It was randomised and double-blind (but perhaps not completely so). Patients were predominantly men with an average age of just under 60 years.

Six hundred and thirty one patients with stable angina and exercise stress tests showing electrocardiogram changes went into a placebo pre-phase. They underwent 48 hour ambulatory electrocardiogram monitoring. The 330 patients with two or more transient ischaemic episodes in 48 hours were randomised to either bisoprolol (4 weeks at 10 mg once daily, followed by 4 weeks of 20 mg once daily) or slow release nifedipine (4 weeks at 20 mg twice daily followed by 4 weeks at 40 mg twice daily). That is, for both treatments there was a low dose phase and a high dose phase.

Two consecutive 24 hour ambulatory electrocardiogram recordings were taken at the end of four weeks (low dose phase) and at the end of eight weeks (high dose phase). All recordings were read blindly in a single laboratory, and only those tapes with at least 75% of the recorded data were used.


There were a number of different outcomes, the main one being total ischaemic burden (ST segment depression in millimetres multiplied by the episode duration in minutes). The number of patients who had episodes of transient ischaemia reduced by 25%, 50%, 75% and 100% was also reported.


The total ischaemic burden dropped (from the same initial level during the placebo phase) by 73% after eight weeks on bisoprolol and by 46% after eight weeks on nifedipine, indicating that ischaemia, measured by ST segment depression and duration, was reduced. The means are shown below.

By using the numbers of patients who had reductions of 25%, 50%, 75% or 100% in the number of transient ischaemic attacks it is possible to calculate NNTs at each level and following low dose phase (4 weeks) and high dose phase (8 weeks) regimens.

Thus at eight weeks the NNT was 2.7 (2.1 to 3.9) for complete cessation of asymptomatic transient ischaemic episodes with 20 mg daily bisoprolol compared with 40 mg twice daily slow release nifedipine in patients with stable angina. This means that one of every three patients with stable angina will have complete cessation of ischaemic attacks with bisoprolol who would not have done so if they were taking nifedipine.

Response and prognosis

A follow up to the TIBBS study [2] looked at consequences one year after the end of the study. The follow-up was of 520 of 545 patients screened, which included the 330 who took part in the study. Information was collected on `hard' events (cardiac and non-cardiac death, nonfatal acute myocardial infarction, hospital admission for unstable angina) and `less hard' events (need for bypass surgery or coronary angioplasty, which are subject to a degree of bias).

The results are interesting because they give a picture of the natural history of angina. The main findings were:-

  1. In patients with more than six ischaemic episodes at baseline, an event occurred in 33%, compared with 25% in those with two to six episodes, and 13% with those with fewer than two episodes.
  2. Hard events were more frequent (12%) in patients with two or more ischaemic episodes, compared with only 5% in those with fewer than two episodes.
  3. Patients with a 100% response rate of transient ischaemic episodes during the TIBBS trial had a 18% event rate at 1 year compared with 32% for non-100% responders.
  4. Patients randomised to bisoprolol had a lower event rate at one year (22%) than those randomised to nifedipine (33%), which may have reflected the influence of reduced ischaemic burden during the trial, or the more frequent use of ß-blockers after the trial stopped in those initially randomised to bisoprolol (47%) than those randomised to nifedipine (32%).

Comparison with nitrates

Two studies [3,4], both from the same group in Groningen, compared placebo with isosorbide dinitrate (ISDN) and bisoprolol in randomised, double bind, crossover comparisons. Once daily 10 mg bisoprolol was compared with 20 mg ISDN three times a day for four or six weeks. A common outcome was the number of angina-free patients during exercise testing at the end of a treatment period.

The L'Abbé plot shows that bisoprolol produced fewer patients with angina attacks. Pooling the results from the two trials, the NNT was 5.0 (2.8 - 21) for 20 mg ISDN three times a day compared with placebo to produce one patient free of exercise induced angina after four to six weeks of treatment. For bisoprolol 10 mg once a day it was 2.8 (1.9 - 5.0).


  1. T von Arnim. Medical treatment to reduce total ischaemic burden: total ischaemic burden bisoprolol study (TIBBS), a multicentre trial comparing bisoprolol and nifedipine. Journal of the American College of Cardiologists 1995 25: 231-8.
  2. T von Arnim. Prognostic significance of transient ischaemic episodes: response to treatment shows improved prognosis. Journal of the American College of Cardiologists 1996 28: 20-4.
  3. MCM Portegies, J Brouwer, L van de Ven, et al. Effects of bisoprolol and isosorbide dinitrate on the circadian distribution of myocardial ischaemia. Current Therapeutic Research 1995 56: 1225-36.
  4. L van de Ven, A Vermuulen, J Tans et al. Which drug to choose for stable angina pectoris: a comparative study between bisoprolol and nitrates. International Journal of Cardiology 1995 47: 217-23.

previous or next story in this issue