Referring to the non-official second listing of the BfArM as a foundation,
there is a database of 32 side effect reports, including those reported by the
Swiss health authorities (IKS), the WHO, as well as those cases described in
the medical literature. The number of 32 reports is of course not synonymous to
32 cases of confirmed causality. The list contains duplicate reports and
questionable classifications. When one collects all the available data for the
individual cases, one arrives inevitably at another evaluation of the data.
2. Duplicate Entries
With a more accurate observation of the case reports listed by the BfArM, one
finds a range of duplicate entries. Obviously, the duplicate entries were not
reconciled against each other when the information was obtained from different
sources. Therefore, the report of one and the same case can lead to a threefold
entry, if, on the one hand, the event is reported directly from the patient,
and on the other hand, the event is also reported via the manufacturer's duty
of notification, and finally, the treating phyisician also turns in a case
report. These redundant data entries are not only difficult to recognize, but
the inflated number of cases has led to a shift in the risk evaluation for a
drug substance. In the case of the BfArM listings, mistakes in data transfer
have exacerbated this situation.
Four, and possibly even five, duplicate reports can be identified in the
analysis of the BfArM listings, as follows:
Under the case numbers 97002825 and 97003551, the BfArM listed a case of
hepatic cell damage attributed to the ingestion of the products
Phyto-Geriatrikum and Eunova. In spite of a deviation in the ages reported for
the patients (one 72 years and the other 75 years), these two cases are
obviously identical: in both cases the patient's initials are „SM“,
in both cases the notification occurred within the same short time frame, and
in both cases the same very unusual combination of medications was reported.
The differences in the ages reported can easily be explained by an error in the
transfer of information from the adverse event registration forms that were
filled out by hand.
Under the case numbers 01001228, 01001924 and 01001928, the listing of the same
event is recorded similarly three times. One of these overlapping cases was
known to the BfArM: the second listing contained a corresponding reference.
This, however, did not prevent the Federal Institute from entering the
excessive number of cases in its
press release regarding the proposed drug safety protocol. Two of the reports
refer to a 38 year old patient with the initials JR or KJ, suffering from
hepatitis, who ingested Laitan in combination with penicillin V. When one
compares the registration forms for these two cases with the third event, it
becomes obvious that they are all referring to the same individual.
A curious duplicate report is recorded in the listing under the case numbers
01003950 and 01003951. Two separate reports for one patient „UW“ are
recorded, wherein UW is reported to have suffered from hepatitis following oral
administration of Kavain Harras N or Kava ratiopharm. In reality, it is a
matter of one and the same patient, with the current case report from the year
2001 and with an earlier case report from the year 1993. These findings should
have led the BfArM to ask some serious questions: In 1993, neither Kavain
Harras N nor Kava ratiopharm were commercially available! An investigation with
the treating physician from the 1993 case showed that the patient's records
did not indicate any identifiable agent responsible for the 1993 occurrence of
liver problems. The 2001 case does indeed report the ingestion of Kava
ratiopharm, however there is considerable doubt of causality from this
medication.
One of the BfArM's listed cases (entry date of July 26, 1999; case number
99006005) may possibly be identical to a case report published in the medical
literature written by Strahl et al. (1998) of acute hepatitis with positive
re-exposure. The external circumstances of these cases, as well as the patient
data, allows one to easily ascertain with some degree of certainty, that they
are referring to the same case, in spite of certain differences in the reports.
Because, ultimately, there is insufficient data available for an accurate
evaluation, both cases are referred to separately in the following analysis.
3. Side Effects With No Relationship to the Ingestion of Kava
In three cases, the ingestion of kava preparations had nothing to do with the
occurrence of hepatic symptoms, among them the case of death mentioned in the
BfArM's press release. One of the cases was not mentioned in the press release
but only appeared in the non-official second BfArM listing. Though the Federal
Institute had not assigned a specific number to this case, it was still
nonetheless listed under the kava side effects.
The case of death described in the press release (BfArM-Number 98004297 dated
June, 4 1998) refers to an 81 year old female patient (WH), who, in the course
of hepatitis, suffered a fatal liver failure. The patient had been taking
Kavatino for over three months at a daily dosage of 60 mg kavalactones. Due to
elevated blood pressure, the patient was also prescribed hydrochlorothiazide as
a co-mediciation at a dosage of 12.5 mg daily. Rare cases of jaundice are known
to occur from this drug, however, just as with kava, it is not known to be a
cause of liver failure. The actual basis for the occurrence of liver symptoms
in this patient may more likely have to do with long-term alcohol abuse, of
which the Institute was indeed aware. The second listing did contain a suitable
reference to this, which, in contrast, was not mentioned in the press release.
The alcoholic genesis of the liver failure was confirmed through a biopsy. The
cirrhotic transformation of the liver had started long before the first
administration of kava preparations began. In any case, the patient died as the
result of her alcohol consumption.
Also, in one suspected case (Number 99005139; dated June 11, 1999), the BfArM
failed, in its press release, to mention the known co-medications. This case
involves a 47 year old female patient, for whom a transient increase in liver
values was recorded with the concomitant ingestion of high dosage fish oil and
Antares. The values normalized without discontinuation of kava therapy.
However, the ingestion of high dosage fish oil is known to produce this
non-pathological reaction pattern. This possible reaction is indeed rare,
however it is stated in the packaging insert of licensed fish oil preparations.
Consequently, a side effect in the liver did not take place.
The third case, reported exclusively in the second listing, goes back to a
report by the firm Schwabe in the context of a 1993 research project. The firm
had informed the BfArM of a 68 year old female patient, who, immediately before
ingesting its product Laitan, had shown elevated liver values. These values did
not worsen during the course of kava therapy. Consequently, there were no
indications of side effects, and this case should have not been included in the
listing.
