Stockpiling oseltamivir: Roche clarifies data for improved mortality with oseltamivir
BMJ 2005; 331 doi: https://doi.org/10.1136/bmj.331.7526.1203-b (Published 17 November 2005) Cite this as: BMJ 2005;331:1203All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
My original submission asked who had briefed Mrs Hewitt on a major
medical issues? 1). I asked because I have to advise a PCT on its ability
to respond to flu pandemic. Having reviewed much of the relevant
evidence, I concluded that:
a) If given within 48 hours of onset of symptoms (as proposed by the
Department of Health) Tamiflu marginally reduces duration of illness.
b) Many patients will feel that this small benefit will offset the
side effects and possible serious harm associated with Tamiflu’s use. As
such when given to a fully consenting patient under “normal” doctor
patient conditions Tamiflu’s use is acceptable
c) Under the Government’s flu pandemic plans, Tamiflu will probably
be distributed- en mass – sometimes from mass distribution centres – in
circumstances, which will make “informed consent” impossible.
d) When public policy expects a citizen to take a medication on
faith, (because it is impractical to obtain informed consent) it is
axiomatic that the benefit to harm ratio must be much higher than in the
normal “doctor-patient” situation.
To my knowledge no such evaluation has been undertaken.
I concluded that there was no place for “mass population Tamiflu” in
our flu pandemic plan. I also believe that until a credible benefit to
harm risk analysis of mass Tamiflu provision has been published, the
medical fraternity should not support such provision.
However, if as Mrs Hewitt stated in October, “Tamiflu will save lives
(during a pandemic)”, then this view may be dangerously wrong and I may
need to reverse it. This is why we all need to know the veracity of Mrs
Hewitt’s statement and the answer to the question, “Who gave her this
advice?”
My second question related to the NNT to avert a flu pandemic death.
Smith and Dutkowski do not answer this(2). I calculate that during a
pandemic, in Birmingham, this NNT probably approaches a million.
Therefore, if each course of Tamiflu costs about £16, the drug cost alone
of each life saved will be about £16 million! Who has advised Mrs Hewitt
that this is a sensible use of NHS money?
1. Stockpiling oseltamivir: What is the number needed to treat with
oseltamivir to prevent one flu death? A Rouse BMJ 2005;331:1203,
doi:10.1136/bmj.331.7526.1203-a
2. Stockpiling oseltamivir: Roche clarifies data for improved mortality
with oseltamivir. James R Smith, Regina Dutkowski BMJ 2005;331:1203,
doi:10.1136/bmj.331.7526.1203-b
Competing interests:
None declared
Competing interests: No competing interests
EDITOR- In response to experts' question, Roche [1] showed very rough
data which, they claimed, supports an improvement in mortality with
oseltamivir. Although I could not have access to the datasets at
<www.eswi.org>, data they showed (32 % and 90 % risk reduction of
pneumonia and death respectively) may not be easily believed by those who
know the limited benefits shown in the randomized controlled trials
(RCTs), potentially harmful effects of the drug shown below and that flu
is a self-limiting disease for previously healthy people.
In the RCT for treatment of adult flu in Japan [2], Kaplan-Meier
curves for proportion of patients without freedom from illness of both
groups were almost completely the same in patients with H3N2 influenza A,
although oseltamivir improved the symptoms of those with H1N1 influenza A
significantly faster than placebo. In the RCT [3] in which chronic asthma
children with influenza were tested, oseltamivir did not improve symptoms
significantly (average duration of illness is rather 14.7 hours less in
placebo group than that in oseltamivir group: p value was not stated).
Oseltamivir did not significantly decrease the proportion of patients with
flu-like symptoms based on data from an RCT for prevention of flu in Japan
(proportion of patients with symptom was 21.9 % in the oseltamivir group
and 23.5 % in the placebo group), although the proportion of patients both
with symptoms and test-proven flu virus was significantly reduced in the
oseltamivir group than in the placebo group[4,5].
Additionally, I have reported three death cases closely related to
the use of oseltamivir recently at a scientific meeting in Japan [6]. One
of the leading Newspaper reported my presentation [7]. Of these only one
case (a 14-year-old male) is included in the 12 pediatric death cases all
from Japan that FDA reported in November 18 [8]. It was concluded that co-
morbidity and confounding factors in many of the cases, limited and
missing data in majority of cases makes it difficult to establish a direct
causal relationship between the use of oseltamivir and the reported deaths
[8].
However, according to my history taking that I presented in Japanese
[6], neither fever nor other drugs could be the causes of his abnormal
behavior and death of a 14-year-old boy. He was one of the two boys
reported by the newspaper [7]. Japanese MHLW (Ministry of Health, Labour,
and Welfare) reported a case of teenage girl as a warning that oseltamivir
could cause hallucination and abnormal behavior [8]. She tried to rush out
of the window as soon as her body temperature went down. Fever was not the
cause of the strange behavior in this case either. This case was not
mentioned in the reports by FDA.
I also reported a previously healthy 2-year-old boy who became breathless
suddenly during afternoon nap and had very low body temperature [6].
Another 2-year-old boy whose parents noticed his abnormal breathing and
took him to a hospital by their car, stopped breathing in the car before
he arrived at the hospital [10]. This case is not included in the 12
Japanese pediatric death cases reported by FDA [8]. These cases strongly
suggest that the infants died from respiratory suppression.
As I mentioned [11], four other sudden infant deaths during sleep
were reported [12]. In total, six infants of sudden death during sleep or
death from abnormal breath were previously healthy and did not take
antipyretics. Causality to such sudden death was considered as follows in
the memorandum of FDA-CDER [13]: "the contribution of the drug to the
death of these patients, especially with the case of sudden death and
cardio-pulmonary arrest, cannot be excluded based upon the information
available."
Previously reported cases of influenza-related encephalopathy (IRE),
especially fatal cases, were closely related to non-steroidal anti-
inflammatory drug (NSAIDs) antipyretics such as diclofenac and/or
mefenamic acid, and the incidence of the fatal cases has decreased since
warnings about NSAIDs antipyretics for flu children were issued in 2000.
Even fatal cases of classical type of IRE did not cause death so suddenly
as the six cases above. So Shiomi called his cases as a new type of IRE
[12]. The new type of IRE cases have started to be reported in the winter
season just after marketing of oseltamivir for children had begun in July
2002.
It should be stressed that the mechanism of sudden death from
respiratory suppression in human infants is confirmed by at least three
animal toxicity studies conducted by Roche [2-4]. Ten minutes to seven
hours (mainly 2 to 3 hours) after the first dose of Tamiflu, many 7-day
old rats died from respiratory suppression, exhibiting spontaneously
decreased movement, weakened respiration which subsequently became
irregular [2-4]. Considering the similarity between the signs and symptoms
in human infants and the juvenile rats, and rather low safety index; only
26 to 40 times based on the peak plasma concentration, close causal
relationship should be suspected between oseltamivir and the infant sudden
death from respiratory suppression during sleep.
Central nervous system suppressants including sedatives, hypnotics
and anesthetics may induce death from respiratory suppression at high
dose, and may induce hallucination, delirium and abnormal behavior by
disinhibition or dyscontrol of the central nervous system at a certain
dose. They are well-known adverse reactions to barbiturate and
benzodiazepins and so on.
It should be noted that FDA did not discuss the animal toxicity
studies indicating suppressive effects of oseltamivir on central nervous
system, the similarity of clinical course of human and animal, and
disinhibition or dyscontrol action of oseltamivir as a central nervous
system suppressant at all.
Number needed to Harm (NNH) of vomiting at the first day of
oseltamivir commencement in the RCT for treatment of children is 15 [3,6].
NNH of headache, nausea and vomiting in the RCT of preventative use of
oseltamivir for adults is 25, 24 and 55 respectively [4,6]. These results
and the evidence of bulged fontanel in the 5 month-old baby adminidtered
oseltamivir for flu prevention [13] indicate that unchanged oseltamivir
alone could increase intracranial pressure even in adults.
If we take into account of these well-known pharmacological actions
of central nervous system suppressants and the effect on intracranial
pressure, full spectrum of psychiatric and neurological adverse effects of
oseltamivir could be well understood.
We should not overlook these potentially harmful effects in addition
to the limited benefit of oseltamivir not only in daily practice but also
in stockpiling oseltamivir.
References
1. James R Smith and Regina Dutkowski Stockpiling oseltamivir: Roche
clarifies data for improved mortality with oseltamivir BMJ, Nov 2005; 331:
1203
2. New drug approval package (NAP) of oseltmivir (in Japanese);
Tamiflu capsule (2000):
http://211.132.8.246/shinyaku/g0012/07/53039900_21200AMY00238.html
3. NAP of oseltmivir (in Japanese); Tamiflu dry syrup (2002):
http://211.132.8.246/shinyaku/g0201/11/5303990_21400AMY00010.html?
4. NAP of oseltmivir (in Japanese); Tamiflu capsule for prevention
(2004) : http://211.132.8.246/shinyaku/g0407/g040703/index.html
5. Kashiwagi S, Kudoh S, Watanabe A, Yoshimura I. Efficacy and safety
of the selective oral neuraminidase inhibitor oseltamivir for prophylaxis
against influenza- -placebo-controlled double-blind multicenter phase III
trial. Kansenshogaku Zasshi (J Infection). 2000 Dec; 74(12): 1062-76 (in
Japanese).
6. Hama R. Discussion of the causal relationship between Oseltamivir
phosphate (Tamiflu), and sudden death and death from abnormal
behavior(presentation at a session of Japanese Society for Pediatric
Infectious Diseases in Tsu, Mie Prefecture November 12 2005) available at:
http://www.npojip.org/sokuho/051112.html (in Japanese) and
http://www.npojip.org/english/no59.html (in English)
7.Mainichi Daily News: Abnormal reaction to Tamiflu medicine tied to
deaths of two boys; http://mdn.mainichi-
msn.co.jp/national/news/20051112p2a00m0na030000c.html
8. Truffa MM. One Year Post-Exclusivity Adverse Event Review
forTamiflu® (oseltamivir)
http://www.fda.gov/ohrms/dockets/ac/05/slides/2005-4180s_03_truffa.ppt#17
9. Japanese FDA MHLW. Pharmaceuticals and Medical Devices Safety
Information No.202 (June 2004; in Japanese )
http://www.mhlw.go.jp/houdou/2004/06/h0624-2/index.html#gai2
10. Fujii F et al. A death case of influenza-related encephalopathy
without showing neurological signs and symptoms. Infection and Immunity in
childhood (2004) 16(2) : 231-232 (in Japanese)
11. Hama R. New type of influenza-related encephalopathy or new
adverse drug reaction? BMJ Rapid Response, 28 February 2005.
http://bmj.bmjjournals.com/cgi/eletters/328/7433/227#98374
12. Shiomi S. Clinical spectrum of influenza-related encephalopathy.
Pediatric internal medicine (in Japanese). 2003: 34(10); 1676-1681.
13. Memorandum of FDA CDER: Tamiflu AE �EReviewed.
http://www.fda.gov/ohrms/dockets/ac/05/briefing/2005-
4180b_06_01_Tamiflu%20AE_reviewed.pdf
Competing interests:
None declared
Competing interests: No competing interests
Although, as Robin Ferner points out, Smith & Dutkowski are
unable to tell the difference between absolute and relative risks, the
papers they reference are available via the internet and do give raw data,
enabling calculation of NNTs.
For example, the Hayden paper (1) Looks at the efficacy of providing
influenza contacts with a 10/7 course of post-exposure prophylaxis with
oseltamivir, compared to just managing the index case with a 5/7 treatment
course of the drug. 62% of index cases had laboratory proven influenza.
Working from their published ITT figures, 40/392 contacts of treated
influenza cases devloped influenza themselves, compared to 11/400 contacts
provided with 10/7 post-exposure prophylaxis - RR 0.27 (95%ci 0.14-0.52),
ARR 0.075 (95%ci 0.04-0.11), NNT of 13.4 (95% ci 9-25).
The abstract of the Kaiser paper (2) gives sufficient detail for NNT
calulations. We are told that 9/1350 oseltamivir-treated influenza
patients were hospitalised compared to 18/1063 placebo patients - an
impressive RR of 0.39 (95% ci 0.18-0.87)... but a not-so impressive NNT of
97 (95%ci 52-726). Then again, at £16 or so for a course of treatment
£1600 to prevent a hospital admission is probably cost-saving.
However, the really interesting thing would be the raw data from the large American cohort study of 176001 people - an NNT to prevent death is a really hard outcome.
And I can't find that on the Roche-backed website, either...
(1) Hayden FG, Belshe R, Villanueva C, Lanno R, Hughes C, Small I, et
al. Management of influenza in households: a prospective, randomized
comparison of oseltamivir treatment with or without postexposure
prophylaxis. J Infect Dis 2004;189: 440-9.[CrossRef][ISI][Medline]
(2) Kaiser L, Wat C, Mills T, Mahoney P, Ward P, Hayden F. Impact of
oseltamivir treatment on influenza-related lower respiratory tract
complications and hospitalizations. Arch Intern Med 2003;163: 1667-
72.[Abstract/Free Full Text]
Competing interests:
None declared
Competing interests: No competing interests
My colleague Andrew Rouse asks what the Number-needed-to-treat (NNT)
might be for oseltamivir. Smith & Dutkowski 'clarify' the data on
improved mortality by quoting relative risk reductions. The web- site
(www.eswi.org) they reference does not give the data needed to calculate
NNTs, either. It is co-sponsored by Roche. Is there something to hide?
Competing interests:
REF provides advice on Therapeutics to Public Health Physicians in Birmingham.
Competing interests: No competing interests
Re: Tamiflu: NNT to prevent a pandemic flu death may be a million
Recent discussions in his journals relating to the stockpiling of
oseltamivir for pandemic influenza have focussed on number needed to treat
(NNT) to avert one fatal case of influenza (1,2). Notwithstanding the
relevance of this parameter, from an economic point of view NNT to avert
one hospitalisation should also be considered. Severe influenza-related
disease leading to hospitalisations has various economic impacts within an
already disruptive society suffering from pandemic influenza. In
particular, hospitalisations will further strain an already overburdened
health-care system during pandemic situations and severe disease involves
production losses in various economic sectors, inclusive crucial civil
services and the health-care system itself.
Whereas NNT to avert one death may be 1800-3200 (3,4) depending on
the exact assumptions, pooled clinical trials (5) show that NNT to avert
one hospitalisation are 97 (all-cause mortality) or 142 (more stringent
influenza-related mortality only). These NNTs for hospitalisations were
derived from pooling 10 randomised clinical trials among patients
presenting to the GP within 36 hours of first ILI-symptom onset (also
influenza infection was confirmed afterwards). As indicated previously
(1), such NNTs probably result in cost savings for oseltamivir treatment
of persons presenting with influenza-like illness at the GP’s office. Next
to these potential cost savings, reductions in hospitalisations and severe
diseases may be crucial in pandemic situations to alleviate the already
overburdened health-care system and avert further breakdown of vital civil
services due to large-scale sickness absence. In particular, it can easily
be derived from the pooled trials that potentially the number of
hospitalisations can be reduced by more than 50% through large-scale
oseltamivir treatment.
Furthermore, we note that additionally indirect effects of
oseltamivir treatment during pandemic situations may occur. In particular,
the probability that an infected person will transmit influenza to others
may be reduced by as much as 80% through antiviral treatment (6).
Consequently, instead of causing 2 new infections on average in a
susceptible population (the so-called R0), less than 1 new infection will
only be caused by an influenza-infected person on oseltamivir treatment.
Next to the direct effect of oseltamivir on hospitalisations outlined
above, also this indirect effect on transmission may prove to be of
crucial importance during an influenza pandemic.
1. Grove ML. NNT for oseltamivir. BMJ.com November 24th
2. Rouse A. Tamiflu: NNT to prevent a pandemic flu death may be a million.
BMJ.com November 28th
3. Thompson WW, Shay DK, Weintraub E, Brammer L, Cox N, Anderson LJ,
Fukuda K. Mortality Associated With Influenza and Respiratory Syncytial
Virus in the United States. J Am Med Assoc 2003;289:179-86
4. Nordstrom et al at www.eswi.org
5. Kaiser L, Wat C, Mills T, Mahoney P, Ward P, Hayden F. Impact of
Oseltamivir Treatment on Influenza-Related Lower Respiratory Tract
Complications and Hospitalizations. Arch Intern Med 2003;163:1667-72
6. Longini IM, Halloran ME, Nizam A, Yang Y. Containing Pandemic Influenza
with Antiviral Agents. Am J Epidemiol 2004;159:623-33
Competing interests:
Dr MJ Postma received research grants from Roche to study the pharmacoeconomics of oseltamivir
Competing interests: No competing interests