Useless and dangerous—fine needle aspiration of hepatic colorectal metastases
BMJ 2004; 328 doi: https://doi.org/10.1136/bmj.328.7438.507 (Published 26 February 2004) Cite this as: BMJ 2004;328:507![Loading Loading](https://www.bmj.com/sites/all/modules/contrib/panels_ajax_tab/images/loading.gif)
All 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.
Dear Sir,
In 1980, a friend of mine who worked in medical research in USA, was
diagnosed, along with two of his colleagues, with early stage prostate
cancer.
We did not know this, until 1995, when there was a campaign here
advising men to get tested for prostate cancer. I discussed this with my
friend.
He said to me "If this test comes back positive, do not let them do a
fine needle aspiration."
I asked "Why?"
He said "Because when they pull the needle out, the suction created
pulls back into the body, along the exit tract, cells, .... which if
cancerous, then have direct access to the rest of the body."
I said to him "I've never read that. How do you know this?"
He replied with word to this effect "It is a matter of debate amongst
my colleagues at the moment. Some of us are extremely concerned about
this. Some say the rate of needle tract metastasis is only around 5 -
10%. I believe that figure is incorrect because it covers all fine needle
aspirations, not just those found to contain malignant cells. If you
don't have cancer cells, you won't get needle tract metastases. I believe
that the rate for those with malignant cancer cells aspirated by the
needle could be around 100%"
He then proceeded to tell me about his two colleagues whose samples
were "malignant", and how they both died very quickly from cancers that
rapidly spread through their bodies.
He chose to do it "his" way. In 1999, after a bout of the flu, he
got pain in the hip, and went to a US hospital, where they told him his
cancer had now spread. He chose hormone treatment only, refusing anything
else they had to offer. He was given three years to live.
He is still alive today.
In this debate above, there are those who say that the study
published is not scientific.
Ever since 1995, I have kept an eye on FNAC of any sort. It seems to
me that the reason there is very little "scientific" evidence as to what
the rates of secondary infection are, as a direct result of this
procedure, is because medical people have assumed that the risks
outweighed the benefits in the ABSENCE of good science.
It also seems to me, that this situation continues to this day
because those who use and advocate this procedure in their practice have
neither the interest or inclination to do proper studies to find out the
answers.
Some might ask why that would be. The answer to that question should
not be that hard to pinpoint....
Sincerely,
Hilary Butler.
Competing interests:
None declared
Competing interests: No competing interests
Useless and dangerous statement about FNAC from liver metastasis from
colo-rectal cancer.
A comment to ”Useless and dangerous fine-needle aspiration of hepatic
colorectal metastasis”,
Metcalfe et al; BMJ vol.328; 28.febr.2004: 507-508
Bjørn Skjoldbye, Christian Nolsoe, Nis Norgaard and Torben Lorentzen
Herlev University Hospital, Copenhagen, DK-2730 Denmark.
E-mail: bos@dadlnet.dk
The title seems improperly tabloid since the content of the paper
does not validate that FNAC is useless or dangerous. Thousands of FNAC
procedures are performed world wide every day with a very low complication
rate. Based on a case story and referring to a material of 51 patients of
whom 5 (10%) had needle tract metastasis after FNAC, the authors
contradict original papers based on thousands of patients (1) and
reported results from other authors presented in reviews (2,3,6) in which
an overall complication rate to interventional ultrasound is approx.
0.20%. A 10% needle tract metastasis rate seem incredible high and should
be subject to a further investigation of the biopsy procedure. If these
figures were valid our hospitals would be filled with patients suffering
of cutaneous metastasis after FNAC - which is not the case.
In a case report the authors describe a patient with a liver metastasis
confirmed by FNAC after radical resection of a Duke C colon tumor. After
hemi-hepatectomy a subcutaneous metastasis develops. The authors do not
present any facts about the biopsy procedure and seem to exclude drainage
ports and the surgical procedure as possible origins. In the discussion
the authors advocate biopsy through laparoscopic ports without discussion
of port-metastasis in connection herewith.
Admittedly, needle tract metastasis do occurs. It is a known, but
rare, complication to FNAC from the liver. And , the issue to substitute
FNAC with tumor characteristics achieved by non-invasive imaging has
been considered before (4). However, imaging diagnostic features of a
liver tumor combined with FNAC provides an unsurpassed diagnostic tool
which is to the benefit to the majority of patients undergoing major
surgery and/or chemo therapy. US-guided FNAC is a safe procedure (1,2,3)
using ultrasound guidance preferably with needle steering device and
performing only one or maximally two needle insertions (5) with a 0.6
(23G) or 0.7 (22G) mm Ciba needle.
PET-CT, MRI and contrast enhanced ultrasound are promising
modalities which may reduce the need for FNAC in the near future. Until
solid evidence of the opposite, FNAC remains a well established tool for
rational patient management before major liver surgery, irradiation and
chemo therapy. US-guided FNAC provides an excellent cost-benefit ratio and
a low complication rate. It is not dangerous and far from useless.
References
1. Nolsoe C, Nielsen L, Torp-Pedersen S, Holm HH. Major complications
and deaths due to interventional ultrasound: A review of 8000 cases. J
Clin Ultrasound 1990;18:179-184
2. Holm HH, Skjoldbye B. Interventional Ultrasound. Ultrasound in Med
& Biol. 1996;vol 22,
No. 7 :773-789
3. Holm HH, Skjoldbye B. Interventional Procedures. Ultrasound in Med
& Biol 1998;vol26,
Suppl.1:131-134
4. Khattar SC, Torp-Pedersen S, Lorentzen T, Nolsoe C, Skjoldbye B,
Court-Payen M, Holm
HH. Liver metastasis: is biopsy verification necessary when sonographic
assessment is certain? European Journal of Ultrasound 1994;1:67-70
5. Skjoldbye B, Horn T, Torp-Pedersen S, Court-Payen M, Khattar SC,
Lorentzen T.
Ultrasound guided fine needle aspiration biopsies from the liver. How
many needle passes?
European Journal of Ultrasound 1996;4:43-47
6. Smith EH. Complication of percutaneous abdominal fine-needle
biopsy. Review. Radiology
1991; 178:253-8
Competing interests:
None declared
Competing interests: No competing interests
In a recent paper a new case of tumour seeding after a
fine needle aspiration cytology (FNAC) performed in a
patients with liver metastasis from colon cancer is
reported (1). On the basis of fatal outcome of this, yet
anedoctal, report authors emphasize the danger of fine
needle biopsy and suggest “that FNAC should be
avoided when hepatic colorectal metastasis are
suspected”. Authors support this advice with the results
of a recent paper (their reference 13) where the
seeding rate after FNAC in 51 patients with liver
metastasis from colon cancer was 10% while it was
1.3% in the other biopsied patients (106 with primary
liver tumours; 46 with non colorectal liver metastasis).
However some weak points should be noted in this
paper:
1. the period of recruitment is very long (1973-1997) so
that in a part of the patients the biopsy was performed
blindly or on the guidance of palpation;
2. the number of needle passes in the lesions is not
reported;
3. reported specificity of FNAC is unacceptably low
(67%);
4. the seeding rate is strongly fluctuating in the various
subgroups of patients, ranging from 0% in the largest
subgroup to 12% in the smallest.
Therefore because of its shortcomings this paper
should not be cited to support recommendations about
the use of FNAC.
Other series cited by Metcalfe and coworkers (their
references 3-7) are related to patients with
hepatocellular carcinoma (HCC), where seeding rate
ranged between 0.6% to 5.1%.
Seeding rate of FNAC (generally considered very low) is
probably underestimated as available data are
retrospective and follow-up of patients is often missing
or insufficient. It is thus particularly interesting to
examine data on complication rate of percutaneous
treatments of liver tumors, radiofrequency thermal
ablation and percutaneous ethanol injection. A recent
paper reporting on complications in 166 patients (114
with HCC and 52 with metastasis, mainly colorectal
one) treated by percutaneous radiofrequency ablation
(expandable system), who all had a prior FN biopsy
(cytology needle or cutting needle). Tumour seeding
was observed in 1 patient with HCC (0.6%) (2); this
seeding rate is similar to that observed in a large
multicenter study with another RF system (cooled
needle electrode calibre) applied in 2320 patients
(1610 with HCC; 683 with metastasis) where 12 (0.5%)
patients had tumour seeding (3), or to the the 0.6%
seeding rate reported in a multicenter study on
complications of percutaneous ethanol injection (PEI)
in 1066 patients with HCC (4); in both these series a
part of the patients had a prior FN biopsy. These
altogether 3552 patients were certainly at higher risk
of seeding in comparison with patients receiving FNAC
alone, both because they received (in a part of cases)
FNAC and percutaneous tumor treatment, and because
in RF treatments large needles are employed which
remain in the tumor for at least 15 minutes; in PEI
treatment a fine needle is used but the puncture is
repeated many times (in the cited paper a mean of 6.7
times).
Keeping thus in mind that for FNAC a lesser seeding
rate should be expected, data on tumor seeding
reported in these papers can be considered a reliable
reference as the patients were carefully followed up for
a period generally long enough to identify this
complication.
Because of overemphasized seeding risk of FNAC
Metcalfe and coworkers suggest to use, instead of
FNAC, a radiologic work up (1) whose sensitivity is
considered about 91% but whose specificity is about
85% (their reference 14), claiming these values are
similar to that of FNAC whereas they are clearly poorer.
FN biopsy has a very high sensitivity and excellent
specificity: in a large study on 2091 FN biopsy either in
HCC and liver metastasis they resulted 91.1% and
99.5% respectively and overall diagnostic accuracy was
93.4% (5). Nowadays there is a tremendous literature
evidence that in experienced hands these results are
easily repeatable. On the other hand the use of a
radiologic work up without biopsy induces in about
1.9% of patients a surgical resection for a benign
lesion (Metcalfe reference 14). This figure is similar to
that found in a previous paper where a radiologic
work-up plus a dosage of alphafetoprotein (AFP) and
carcinoembrionic antigen (CEA) without liver biopsy
were performed in 160 patients with liver lesions. In
that series 4 patients (2.5%) with non malignant lesion
were surgically treated (6). We deem hardly acceptable
to expose 1.9%-2.5% of subjects with benign liver
lesions to a risk of liver resection to avoid a tumor
seeding rate probably less than 0.5-0.6%. However we
would point out that a non-invasive diagnostic approach
has to be preferred whenever possible and reliable, as
it has been advised for hepatocellular carcinoma: for a
tumor greater than 2 cm, two positive imaging
techniques or one positive imaging technique and AFP
values higher than 400 ng/ml suffice for diagnosis (7).
Debate on seeding risk deserve also some technical
considerations: the use of laparoscopic biopsy
suggested by Metcalfe when a pathologic diagnosis is
mandatory does not eliminate the seeding risk and
introduces the risk of port insertion; in patients eligible
to a curative resection FNAC can be instead performed
by using a coaxial technique, which can eliminate
seeding risk. It is well known that number of needle
passes can influence tumor seeding rate: for this
reason we have proposed the use of an immediate
staining (8) to reduce the number of needle passes.
Another factor influencing cell dragging during biopsy is
irregularity of needle surface, and we have thus
suggested to avoid needles with the so-called echo
markers (9). Track for FNAC of a liver lesion should be
cautiously chosen so as to traverse normal adjacent
liver parenchyma prior to entering a metastatic deposit,
especially if a peripherally-based one, to reduce the
likelihood of malignant cutaneous seeding.
Finally, a clinical remark: it should be recalled that
tumour implantation in more than half of reported
cases did not influence patient outcome.
We therefore think statements of Metcalfe and
coworkers clashes against literature evidences: FNAC
of liver tumors is a powerful diagnostic tool, which, as
every invasive test, should be restricted to situations in
which its result have an impact on patient
management; FNAC is certainly safe when performed
with appropriate operating procedures based on the
sound principles of biopsy technique.
References
1. Metcalfe MS, Bridgewater FH, Mullin EJ, Maddern GJ
Useless and dangerous—fine needle aspiration of
hepatic colorectal metastases BMJ 2004; 328:
507-508
2. Buscarini E, Buscarini L Radiofrequency thermal
ablation with expandable needle of focal liver
malignancies: complication report Eur Radiol
2004;14:31-37
3. Livraghi T. Solbiati L, Meloni F, Scott-Gazelle G,
Halpern EF, Goldberg SN Treatment of focal liver
tumors with percutaneous radio-frequency ablation:
complications encountered in a multicenter study.
Radiology 2003; 226:441–45
4. Di Stasi M, Buscarini L, Livraghi T, Giorgio A, Salmi A,
De Sio I, Brunello F, Solmi L, Caturelli E, Magnolfi F,
Caremani M, Filice C Percutaneous ethanol injection in
the treatment of hepatocellular carcinoma. A
multicenter survey of evaluation practices and
complication rates. Scand J Gastroenterol 1997; 32:
1168-1173
5. Buscarini L, Fornari F, Bolondi L, Colombo P,
Livraghi T, Magnolfi F, Rapaccini GL, Salmi A
Ultrasound-guided fine-needle biopsy of focal liver
lesions: techniques, diagnostic accuracy and
complications. A retrospective study on 2091 biopsies.
J Hepatol 1990; 11: 344-348
6. Torzilli G, Minagawa M, Takayama T, Inoue K, Hui AM,
Kubota K, Ohtomo K, Majuuchi M Accurate preoperative
evaluation of liver mass lesions without fine-needle
biopsy Hepatology 1999; 30: 889-893
7. Bruix J, Sherman M, Llovet JM, et al. Clinical
management of hepatocellular carcinoma.
Conclusions of the Barcelona-2000 EASL conference. J
Hepatol 2001; 35: 421-30
8. Civardi G, Fornari F, Cavanna L, Di Stasi M, Sbolli G,
Buscarini L Value of rapid staining and assessment of
ultrasound guided aspiration biopsies. Acta Cytol 1988;
32: 552-554
9. Buscarini E, Foroni R, Rossi S, Di Stasi M, Silva M,
Marinone G, Degli Antoni G, Buscarini L Fine needles
with echo markers: increasing cell dragging during
biopsy. Acta Cytol 1997; 41: 1246-1249
Competing interests:
None declared
Competing interests: No competing interests
Well the damage is done now. Thank you BMJ. No amount of evidence
based, reasoned debate is likely to undo the damage that your Grubb-street
one-liner has caused to an inexpensive, relatively non-invasive and often
very useful technique.
I would suggest that the only thing that is useless and dangerous
here is an inappropriate title attached to a rather slight case report.
If the title is the authors (it does sound a bit surgical) then you should
have changed it. If it is yours then you should be ashamed.
An exchange of correspondence buried in a subsequent edition of your
tabloid will not fix this one. A reflective editorial on the importance
of maintaining standards in medical journalism might. Useless and
dangerous?? I should say so.
Simon Knowles
Competing interests:
I'm a cytopathologist.
Competing interests: No competing interests
the author states 'if a hepatic lesion is discovered at the same time
or after a primary colorectal malignancy is diagnosed, and if it appears
to be malignant either on imaging or on appearences at laparotomy, then it
should be treated as such, without biopsy'.
I would like to ask, if he commences anti-cancer therapy
for presumed liver metastasis that do not turn out to be malignant, then
how could he defend himself. I would say that histological confirmation is
crucial before committing the patient to any form of cancer therapy.
The author also mentions that in equivocal situations were it is
difficult to be certain about the nature of the liver lesion, then a three
months wait is recommended followed by repeat imaging.
I would like to state that it is not fair to keep patients waiting
for three months in such situations before a diagnosis is confirmed. And
what if the liver lesions have not changed on repeat imaging, would he
wait another three months.
If the author is against FNAC for the reasons he mentioned,then the
answer is to perform another procedure for histological confirmation, such
as laparoscopic biopsy, and not to do without histology.
regards
A Haydar
Competing interests:
None declared
Competing interests: No competing interests
It is most surprising that the BMJ has decided to publish the article
with a title, eye catching as it might be, but smacks of tabloid
journalism. To dismiss outright as useless and dangerous the procedure of
FNAC of suspected colo rectal meatstases will raise doubts in the minds of
many clinicians of the safety of this procedure not only in colorectal
metastases but in any liver metastases. It certainly raises medico legal
implications for those who perform FNAC of liver metastases.
There have been recent examples of chaos and confusion caused in the
minds of the public and the medical profession resulting from publications
based on inadequate, inaccurate or biased data.
I hope the BMJ will show some restraint in the use of attention
seeking titles till there has been more debate on the subject. There are
many of us in this country who have peroformed several hundreds of FNAC
and percutaneous biopsies without encountering so many complicaions from
seeding. Admittedly we may not have been aware of our own complication
rates and I shall be strenuously attempting to gather data from my own
series and look forward to making it public.
The views of the oncologists need to be canvassed as to how essential
it is for them to have cytological or histological diagnosis specially
with the increasing possibility of targeted chemotherapy. Many FNAC are
performed to provide the information for the oncologist. CT, MR, US, PET
cannot provide a conclusive histological diagnosis in many instances.
Clearly tissue characterisation using these techniques may be a solution
but is in the distant future.
The authors do not raise a great deal of confidence in that the
management pathway reported leaves something to be desired if it is not a
case of negligence. Having found a tumour which given the circumstances
might have been suggestive of colorectal metastasis there is no evidence
that there had been a search for a primary site at the time of the
ultasound examination.
Competing interests:
I have receently been granted patents in the United States and in Europe for enhanced visualisation of devices for ultrasound guided interventional work and are in the commercialisation phase of this invention.
Competing interests: No competing interests
The authors underline the risks of fine needle aspiration cytology
(FNAC) for hepatic colorectal metastases. But their proposed
contraindication of FNAC and the possible diagnostic alternatives for
liver lesions without apparent primary malignancy are not evidence-based.
For instance, in suspected metastases of unknown origin, they recommend
that “the investigations (…) should be directed to detect the primary
lesion”, something clearly opposed to nowadays guidelines for the
diagnostic work up of unknown primary tumours that limit diagnostic
approaches. Routine use of invasive surgery for the diagnosis of liver
lesions implies a risk of procedure-associated morbidity and morbidity
that may be not reasonable in some patients with advanced cancer. Positron
emission tomography can not be considered the gold standard to
differentiate between benign and malignant lesions as its sensitivity is
not significantly different to that of FNAC.
And, finally, we suspect that
even the authors will not accept for themselves their proposed “natural
selection” criterion consisting in a three months delay to confirm that a
suspected benign (and resectable) lesion has become unresectable and
fatal.
Competing interests:
None declared
Competing interests: No competing interests
This single case does highlight the risks of a FNA procedure and is a
stimulus for followup of FNAC procedures at our hospital. A cytological
diagnosis was obtained in this case, albeit with sequelae.
Rather than a plethora of tests(PET or laparascopic biopsy)which
respectively cannot give definitive histology or require a GA, FNA of the
liver lesion does provide a less invasive single-step cytological
diagnosis.
The use of coaxial technique with FNAC may reduce risk of tumour
seeding to that of laparoscopy biopsy.
What is the documented rate of abdominal wall metastsases with
laparoscopic access? Morbidity due to port insertion must also be
considered.
How often are there cytological differences between liver metastases
and the primary colonic lesion? There was difference in response of the
liver and abdominal wall metastasis. Was there a post treatment
effect/further differentiation due to the chemotherapy making the
abdominal wall lesion more difficult to treat?
My concerns are that although the primary tumour treatment was
effective and there was a complication fom a single diagnostic test, we
are being told that complex tests should replace less complex tests,
without regard to costs of these tests. The specific tumour histology
could have also been discussed further.
Radiology is an effective modality that reflects pathology, but it
always needs pathological and clinical correlation.
If there is a 3% risk of abdominal wall seeding, co-axial technique
may be more cost effective than the PET or laparoscopic techniques that
have been proposed.
Competing interests:
None declared
Competing interests: No competing interests
Biopsy of potentially operable hepatic colorectal metastases-not useless but certainly dangerous.
We have followed with interest the recent debate regarding tumour
seeding in the aftermath of fine needle aspiration cytology (FNAC) in
patients with potentially resectable hepatic colorectal liver metastases
(1). Metcalfe and colleagues’ verdict of “useless and dangerous” seems
to have provoked strong emotions amongst some of your readers, and we
should like to contribute two observations.
Our staging protocol comprises liver-specific MRI, chest CT and the
selective use of PET, laparoscopy and / or a “trial of time”, but
excluding biopsy. Since 1986 we have undertaken more than a thousand liver
resections for metastatic cancer without resort to pre-operative biopsy or
FNAC with only seven ‘false positives’. In two patients, hepatic cysts
were diagnosed at operation and resection was deferred, whilst liver
resection was undertaken without complication in the other five (three
haemangiomas and two cysts).
Recent analysis of 598 consecutive patients undergoing radical
resection of colorectal liver metastases examined specifically the 90
patients in whom diagnostic biopsy had been performed prior to referral
(2). Histologically-confirmed tumour seeding at the site of biopsy was
confirmed in 17 patients (19%). This concurs with the findings of another
two recent studies (3,4). In every patient in our series, these chest and
abdominal wall deposits were excised at the time of liver resection.
Nevertheless, our analysis showed that survival after liver resection was
substantially and significantly diminished compared to well-matched
patients in whom no biopsy or FNAC had been attempted (5).
In our experience, the non-invasive evaluation of potentially
resectable colorectal liver metastases is at least 99% specific.
Furthermore, the violation of tissue planes by biopsy or FNAC
significantly compromises patient survival. We believe, therefore, that
Metcalfe and colleagues’ choice of title is apt. Consultation with a
specialist hepatobiliary surgical team is recommended before a ‘tissue
diagnosis’ is attempted in such patients.
1. Metcalfe MS, Bridgewater FHG, Mullin EF, Maddern GJ. Useless and
dangerous-fine needle aspiration of hepatic colorectal metastases. BMJ
2004; 328:507-8.
2. Jones OM, Rees M, John TG, Bygrave S, Plant G. Resectable
colorectal liver metastases: to biopsy or not to biopsy? Colorectal
Disease 2004; 6:1-34.
3. Rodgers MS, Collinson R, Desai S, Stubbs RS, McCall JL. Risk of
dissemination with biopsy of colorectal liver metastases. Dis Colon Rectum
2003; 46:454-8.
4. Ohlsson B, Nilsson J, Stenram U, Akerman M, Tranberg KG.
Percutaneous fine-needle aspiration cytology in the diagnosis and
management of liver tumours. Br J Surg 2002;89:757-62.
5. Jones OM, Rees M, John TG, Bygrave S, Plant G. Biopsy of
colorectal metastases causes tumour dissemination and adversely affects
survival following liver resection. Br J Surg 2004 (in press).
Competing interests:
None declared
Competing interests: No competing interests