Adverse drug reactions as cause of admission to hospital: prospective analysis of 18 820 patients

The comments by Dr. Firenzuoli et al are well taken.
Too many self-styled 'healers' are active and many do not have any
training, let alone even minimal knowledge of what they are doing.

However, there are large numbers of properly trained practitioners in all
parts of the globe who fill a void that otherwise would not even be
recognised.

Chiropractors and Osteopaths receive training comparable to medical
doctors, so do Naturopaths in some countries.
A quick glance at malpractice insurance premiums will give some insight
into the real risks craeted by alternative practitioners, as judged by the
insurance industry.

The comments in this letter talk of vertebral manipulation with its 'real'
risk of cerebrovascular accidents but fail to mention that, to date, the
vast majority of the very small number of CVA victims suffered their fate
at the inexperienced hands of medical practitioners 'practicing'
manipulation technique.
"Herbal drugs can be hepatotoxic, nephrotoxic, cardiotoxic...." - a
statement which is undoubtedly true but is a bit rich, coming as it does
from a member of the establishment whose track record is not much better
than that of a medieval battlefield.
After all, almost every drug fits the description of ---toxic.

Then, it is stated that the placement of a medical doctor (experienced in
complementary therapies) became necessary and that, so far, 65 'main
adverse events' had occured, of which 27 requested hospital admission.

Well, compare this to the many hundreds of thousands who fall victim
to the number one cause of death today: Allopathic Medicine.
Another thought is the fact that, in the USA, visits to unorthodox
practitioners at times exceed those to the orthodox doctors.
This means, especially when one considers the necessary personal cash
outlays for unorthodox treatment, that vast numbers of patients are
unhappy with their allopathic treatment.
Probably a satisfactory solution is in place in some central European
countries, where 'properly trained' medical doctors have chosen to
practice complementary medicine.
I could name a few dozen names off the top of my head but prefer to spare
those practitioners the additional attacks from their -often jealous and
backstabbing- colleagues.

Competing interests:
None declared

The study of Pirmohamed and colleagues (1) provides valuable new
information on adverse drug reactions (ADRs) as causes of hospitalisation
and the associated burden on the NHS. It is interesting that, among
individual drugs, warfarin was the second most common cause of such
admissions (129/1225, 10.5%), all of which involved over-anticoagulation
and bleeding events that were occasionally fatal. This is consistent with
the results of our recent audit of warfarin-related bleeding events among
practitioner-led inpatient and outpatient anticoagulation therapy at two
UK hospitals (2). In the latter study a total of 106 hospital admissions
(0.6% of all acute admissions) over a 10 month period were due to such
events, some of which proved fatal. These findings are of concern for
healthcare professionals involved with the management of patients
receiving anticoagulation therapy with warfarin.

The authors highlight drug interactions as one possible cause of ADR-
related hospitalisations, and this is especially relevant to warfarin
therapy (3). However, other forms of warfarin interaction, such as with
food, alcohol and herbal products, can also contribute to significant
difficulties with anticoagulation control. Drug interactions with warfarin
that give rise to decreased efficacy of anticoagulation may also culminate
in adverse outcomes necessitating hospitalisation (e.g. stroke in a
patient receiving anticoagulation therapy for atrial fibrillation). Such
outcomes would not necessarily have been captured in Pirmohamed and
colleagues’ study. Of possibly greater significance are the associated
indirect costs of such outcomes, not the least of which includes the
detrimental cost to a patient’s quality of life (4,5).

With regard to warfarin, therefore, the study of Pirmohamed and
colleagues highlights the need for increased awareness of interactions and
other problems associated with the use of this agent that may lead to
difficulties with anticoagulation control (and, in turn, increased risk of
hospitalisation) and requirement for greater frequency of monitoring.
Warfarin is also the third highest subject of dispensing errors in the UK,
given the availability of numerous doses (6). Clearly, new fixed-dose
anticoagulants that are not subject to the numerous limitations of
warfarin, and which have a better risk–benefit profile and lack of
interactions with food, alcohol and other medications, would be
therapeutically advantageous and would present less of a concern regarding
dosing errors. Such agents should significantly reduce the burden on the
NHS of ADRs arising from anticoagulation therapy with warfarin.

References

1. Pirmohamed M, James S, Meakin S, Green C, Scott AK, Walley TJ, et
al. Adverse drug reactions as cause of admission to hospital: prospective
analysis of 18 820 patients. BMJ 2004;329:15–9.

2. Rhodes S, Green ES, Bond S, James R, Taylor M, Rose P, et al.
Bleeding complications of oral anticoagulant therapy: a prospective audit.
Br J Haematol 2004;125 Suppl. 1:57.

3. Harder S,.Thürmann P. Clinically important drug interactions with
anticoagulants: an update. Clin Pharmacokinet 1996;30:416–44.

4. Clarke P, Marshall V, Black SE, Colantonio A. Well-being after
stroke in Canadian seniors: findings from the Canadian Study of Health and
Aging. Stroke 2002;33:1016–21.

5. Lai SM, Studenski S, Duncan PW, Perera S. Persisting consequences
of stroke measured by the Stroke Impact Scale. Stroke 2002;33:1840–4.

6. Department of Health. Building a safer NHS for patients: improving
medication safety. A report by the Chief Pharmaceutical Officer, 22
January 2004. Available at: www.dh.gov.uk/Home/fs/en

Competing interests:
Sarah Green has received reimbursement for attending symposia from AstraZeneca, Leo Pharmaceuticals and Aventis. Sue Rhodes and Sarah Bond have received funding from AstraZeneca, Leo Pharmaceuticals, Pharmacia (now Pfizer) and Roche Diagnostics to attend symposia over the past 5 years. AstraZeneca and Leo Pharmaceuticals have also paid for posters to be prepared which they have presented at various symposia although we received no fee for presenting the work. AstraZeneca funded the development of the audit proforma and database to collate, analyse and cost adverse outcomes in anticoagulant therapy. Peter Rose has received reimbursement for attending symposia from AstraZeneca, Leo Pharmaceuticals, Sanofi Synthelabo and has received a fee for speaking at a symposium. He edits Thrombus, a journal published by Heywood Medical, which receives an educational grant from Leo Pharmaceuticals. There is nothing else to declare.

Do the shortcomings of clinical trials or the lack of advance in
pharmacogenetics really excuse the disconcertingly high incidence of
medicines-related adverse effects? According to Pirmohamed and
colleagues' study of admissions to Merseyside hospitals, nearly three-
quarters of adverse drug reactions were avoidable; the adverse effects
were well recognised and predictable.

Of course we must do everything possible to predict more accurately
the nature and incidence of adverse effects. However, to make a
difference to hospital admissions, it is also important to use the
knowledge we already possess more effectively. How can we achieve this?

Firstly, the healthcare professional needs clear, reliable and
relevant information to inform decisions on prescribing, dispensing, and
administration. Such information needs to address both the practical
needs (How often should the medicine be taken?) and decision-making needs
(What's the most appropriate medicine? What relative and absolute contra-
indications are relevant for my patient? How can the risk of adverse
reactions be minimised?). The BNF is designed to answer these very
questions.

The clinician is overwhelmed by a vast quantity of information with
little time to review the evolving knowledge. However, the BNF processes
all relevant sources of information and casts the knowledge in a form that
is useful in clinical practice.

Secondly, students training as healthcare professionals should:

- be educated in the principles of good prescribing;

- use essential drug information sources competently;

- understand how adverse effects of drugs can be minimised;

- recognise their future role in reporting adverse drug reactions.

Thirdly, patients need information on side-effects that is relevant
and helpful rather than overwhelming and frightening. The BNF is keeping
up with the welcomed trend of showing the frequency of side-effects.
Although conveying the frequencies of adverse effects to patients remains
a challenge, clinicians will at least know which side-effects are most
likely to occur.

Competing interests:
None declared

In the largest prospective survey in the United Kingdom(1) of adverse
drug reactions (ADRs) it was reported that up to 6.5% of all Hospital
admissions are related to ADRs and that over of 2% of patients admitted
with ADR died, suggesting that adverse effects may be responsible for the
death of 0.15% of all admissions.

One type of ADR is allergic or anaphylactic reaction. Fatalities as a
result of drug-induced anaphylaxis have been reported in several studies.

In a recent study(2)contacted again in the United Kingdom it was found
that in the interval from 1992 through 1998, 168 fatalities were
anaphylactically induced and 39% of these were drug induced.
Kounis Syndrome(3),(4),(5), is the concurrence of allergic or anaphylactic
reactions with acute coronary syndromes leading to allergic angina and/or
to allergic myocardial infarction. Drugs can induce Kounis syndrome and
there are reports showing that several categories of drugs are capable
of inducing this syndrome as follows:

1. Antibiotics: ampicillin,ampicillin/sulbactam, amoxicillin, amikacin,
cefazolin, cefoxitin, cefuroxim, cephradine, cinoxacin, lincomycin,
penicillin, sulbactam/cefoperazone, sulperazon,vancomycin.

2. Analgesics: dipyrone.

3. Antineoplastics: 5-flouorouracil, capecitabine, carboplatin.

4. Contrast media: iohexole, loxaglate, meglumine diatrizoate, sodium
indigotindisulfonate.

5. Corticosteroids: betamethasone, hydrocortisone.

6. Intravenous anaesthetics: etomidate, rocuronium bromide, suxamethonium,
trimethaphan.

7. NSAIDs: diclofenac, naproxen.

8. Skin disinfectants: chlorhexidine, povidone-iodine

9. Thrombolytics and anticoagulants: heparin, lepirudin, streptokinase,
urokinase.

10. Others: allopurinol, enalapril,esmolol, dextran-40, fructose, insulin,
iodine, protamine, tetanous antitoxin, glaphenine.
With the above plethora of drugs in the medical armamentarium and with so
many patients it is strange that in the above excellent survey(1) none of
these drugs was identified to cause any reaction.

1. Pirmohamed M, James S, Meakin S, green C, Scott AK, Walley TJ et
al. Adverse drug reactions as cause of admission to hospital: prospective
analysis of 18820 patients. Br Med J 2004; 329: 15-19

2. Pumphrey R. Lessons for management of anaphylaxis from a study of fatal
reactions. Clin Exper Allergy 2000; 30: 1144-1150.

3. Zavras GM, Papadaki PJ, Kokkinis SE, Kalokairinou K, kouni SN,
Batsolaki M, et al. Kounis syndrome secondary to allergic reaction
following shellfish ingestion. Int J Clin Pract 2003; 57: 622-624.

4. Koutsojannis CM, Kounis NG. Lepirudin anaphylaxis and Kounis syndrome.
Circulation 2004; 109: e315.

5. Electronic responses. Drug induced hypersensitivity syndrome and kounis
syndrome. bmj.bmjjournals.com 2004.
bmj.bmjjournals.com/cgi/eletters/328/7451/1292.

Competing interests:
None declared