PostScript

of colorectal cancer. The main findings from colorectal cancer globally each year.2 Published Online First 24 December 2020
showed that aspirin, non-­ steroidal anti-­ Nonetheless, it is imperative that studies
Gut 2021;70:2215–2216. doi:10.1136/

Gut: first published as 10.1136/gutjnl-2020-323731 on 5 January 2021. Downloaded from http://gut.bmj.com/ on December 15, 2024 by guest. Protected by copyright.
inflammatory drugs, magnesium, folate report absolute risk estimates with relative gutjnl-2020-323505
and high consumption of fruits, vegeta- risk estimates to permit a more meaningful
bles, fibre and dairy products were associ- interpretation of the evidence. ORCID iD
ated with decreased incidence of colorectal Fruit and vegetable consumption Blake J Lawrence http://​orcid.​org/​0000-​0002-​8772-​
8226
cancer. Whereas frequent consumption of was reported as the largest protective
alcohol and meat were associated with effect, potentially decreasing risk of
increased incidence of colorectal cancer. colorectal cancer, whereas consumption References
1 Chapelle N, Martel M, Toes-­Zoutendijk E, et al.
Chapelle et al also reported the quality of meat was reported as the largest effect Recent advances in clinical practice: colorectal cancer
of evidence as overwhelmingly very low potentially increasing risk of colorectal chemoprevention in the average-­risk population. Gut
to low and concluded their review by cancer. All other relative risks reported 2020;69:2244–55.
suggesting that these findings will assist by Chapelle et al will likely translate 2 Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer
incidence and mortality worldwide: sources, methods
clinicians when advising average risk into absolute risk estimates smaller than and major patterns in GLOBOCAN 2012. Int J Cancer
patients, yet they did not report the statis- the results reported above. Further- 2015;136:E359–86.
tical result most relevant to an average risk more, these small absolute risks are 3 Holmberg MJ, Andersen LW. Estimating risk ratios
patient living in the community: absolute calculated from meta-­analyses of obser- and risk differences: alternatives to odds ratios. JAMA
2020;324:1098–9.
risk. vational studies classified as very low
4 Noordzij M, van Diepen M, Caskey FC, et al. Relative risk
Absolute risk is the estimate of the like- to low quality evidence. As expressed versus absolute risk: one cannot be interpreted without
lihood of the occurrence of an outcome. by others,4 5 caution must be used by the other. Nephrol Dial Transplant 2017;32:ii13–18.
For example, the estimated global abso- clinicians when counselling average 5 Ioannidis JPA. Implausible results in human nutrition
lute risk of colorectal cancer incidence risk patients based on the findings from research. BMJ 2013;347:f6698.
is 2%.2 When examining whether medi- studies solely reporting relative risks
cations and lifestyle factors are associ- from low quality observational studies,
ated with increased or decreased risk of and without reporting the associated Where should ascitic drains be
disease incidence (in this case, colorectal absolute risks. To ensure researchers, placed? Revisiting anatomical
cancer) it is important that research find- clinicians and patients are provided with
ings are reported within the context of the most reliable and accurate evidence,
landmarks for paracentesis
the overall absolute risk of disease inci- absolute risk (in addition to relative risk)
dence.3 4 Research findings primarily must be reported in future clinical trials, We welcome the recent publication of the
reported as relative risk associations (such systematic reviews and meta-­analyses. BSG/BASL guidelines on the management
as those by Chapelle et al) provide an esti- of ascites in cirrhosis which will serve as a
mate of the difference in risk of colorectal Blake J Lawrence ‍ ‍,1,2 Elise Alexander,1,2 framework for patient management across
cancer incidence between the groups of Hayley Grant,1,2 Moira O'Connor1,2 the world.1 Following clinician feedback
patients included in each study. However, 1
WA Cancer Prevention Research Unit, Curtin on Twitter,2 we wish to respectfully high-
due to the way in which relative risks are University, Perth, WA, Australia light concerns regarding their guidance on
2
calculated they must first be transformed School of Psychology, Curtin University, Perth, WA, landmarks for therapeutic paracentesis.
into absolute risk estimates before the Australia Our first concern relates to the descrip-
potential benefit or harm of medications Correspondence to Dr Blake J Lawrence, School tion of the landmarks. The authors recom-
and lifestyle factors associated with risk of Psychology, Curtin University, Perth, WA 6102, mend this to be ‘at least 8 cm (laterally)
of colorectal cancer (for an average risk Australia; ​blake.​lawrence@​curtin.​edu.​au from the midline and 5 cm above the
population) can be communicated to Twitter Blake J Lawrence @blakej_lawrence symphysis (pubis)’.1 This was predicated
researchers, clinicians and patients. Contributors All authors (BJL, EA, HG and MO) on three studies (two laparoscopic; one
For example, Chapelle et al reported contributed to the preparation and revision of this cadaveric) on the anatomical course of
consumption of fruits and vegetables as manuscript. the inferior epigastric artery.3–5 While
associated with up to a 49% relative risk Funding The authors have not declared a specific this approach avoids puncturing the
reduction in the incidence of colorectal grant for this research from any funding agency in the vessel, there remains a risk of injury to
public, commercial or not-­for-­profit sectors.
cancer, RR=0.51 (95% CI 0.19 to 1.32). underlying solid organs which can result
However, a 49% reduction in relative risk Competing interests None declared. in haemorrhage or perforation. Of note,
converts to approximately a 1% reduction Patient and public involvement Patients and/or these studies may not be generalisable to
in the absolute risk of colorectal cancer in the public were not involved in the design, or conduct, patients with distortion of the abdominal
or reporting, or dissemination plans of this research.
the broader population (eg, absolute risk cavity due to gross ascites.
of 2% * 0.49=0.98%). A similar result Patient consent for publication Not required. Second, the figures depicting the land-
can be found for consumption of meat, Provenance and peer review Not commissioned; marks for paracentesis require clarifi-
where Chapelle et al reported up to a externally peer reviewed. cation. In Figure 4A of the published
25% relative risk increase in the incidence © Author(s) (or their employer(s)) 2021. No commercial paper,1 the umbilicus appears closer to the
of colorectal cancer, RR=1.25 (95% CI re-­use. See rights and permissions. Published by BMJ. subcostal plane than the symphysis pubis;
1.15 to 1.36). This estimate converts into this is not anatomically representative.
a 0.5% increase in absolute risk (ie, 2% The safe zones (denoted in green) appear
* 1.25=2.5%). It is important to note, To cite Lawrence BJ, Alexander E, Grant H, et al. Gut too superior and lateral; these exceed the
however, that what may be perceived as a 2021;70:2215–2216. ‘5 cm above the symphysis’ description
relatively small 1% reduction in absolute Received 30 October 2020 which is ambiguous, and should stipulate
risk still equates to approximately 680 000 Revised 10 December 2020 an upper bound distance. In Figure 4B of
fewer cases and 347 000 fewer deaths Accepted 13 December 2020 the published paper,1 the paracentesis site
2216 Gut November 2021 Vol 70 No 11
 PostScript
Patient and public involvement Patients and/or
the public were not involved in the design, or conduct,
or reporting, or dissemination plans of this research.

Gut: first published as 10.1136/gutjnl-2020-323731 on 5 January 2021. Downloaded from http://gut.bmj.com/ on December 15, 2024 by guest. Protected by copyright.
Patient consent for publication Not required.
Provenance and peer review Not commissioned;
externally peer reviewed.
© Author(s) (or their employer(s)) 2021. No commercial
re-­use. See rights and permissions. Published by BMJ.

To cite Siau K, Robson N, Bollipo S, et al. Gut

2021;70:2216–2217.
Received 28 November 2020
Revised 16 December 2020
Accepted 16 December 2020
Published Online First 5 January 2021

Gut 2021;70:2216–2217. doi:10.1136/

gutjnl-2020-323731

Figure 1 Anatomical landmarks for paracentesis. ASIS, anterior superior iliac spine. ORCID iD
Keith Siau http://​orcid.​org/​0000-​0002-​1273-​9561

photographed corresponds to the diagram The contralateral (left) McBurney’s point References
but appears too superior and lateral to be is favoured, as the abdominal wall here 1 Aithal GP, Palaniyappan N, China L, et al. Guidelines
considered safe without ultrasound guid- is thinner, with deeper ascitic pool and on the management of ascites in cirrhosis. Gut
2021;70:9–29.
ance, as this carries a risk of splenic injury lower theoretical risk of perforation as 2 Siau K. @drkeithsiau Guidelines on the management
and bowel perforation. the sigmoid colon is more mobile than of ascites in cirrhosis. Twitter, 2020. Available: https://​
Patient safety must always be prioritised. the fixed caecum.9 Care should be taken twitter.c​ om/​drkeithsiau/​status/​1319682365082533890
Ultrasound guidance mitigates the risks to avoid far lateral sites, engorged veins [Accessed 16 Nov 2020].
3 Hurd WW, Bude RO, DeLancey JO, et al. The location
of paracentesis-­ related complications, or previous scars.7 The landmarks can
of abdominal wall blood vessels in relationship to
but its availability is limited in resource-­ be shifted slightly laterally to account abdominal landmarks apparent at laparoscopy. Am J
poor settings and in some UK hospitals. for scarring from regular paracentesis. Obstet Gynecol 1994;171:642–6.
Training in bedside ultrasound is also We believe this approach to be safer and 4 Saber AA, Meslemani AM, Davis R, et al. Safety zones
variable. Moreover, with the COVID-19 more anatomically correct in the absence for anterior abdominal wall entry during laparoscopy:
a CT scan mapping of epigastric vessels. Ann Surg
pandemic, transporting ultrasound of ultrasound guidance, and encourage a 2004;239:182–5.
machines between patients and across revision to Figure 4. However, this caveat 5 Joy P, Prithishkumar IJ, Isaac B. Clinical anatomy of
sites risks transmission. Thus, it remains should by no means detract from the the inferior epigastric artery with special relevance to
important to teach paracentesis landmarks commendable efforts by the authors in invasive procedures of the anterior abdominal wall. J
Minim Access Surg 2017;13:18–21.
in a clear and safe manner without sole formulating these guidelines.
6 Lee SY, Pormento JG, Koong HN. Abdominal
reliance on ultrasound. paracentesis and thoracocentesis. Surg Laparosc Endosc
Keith Siau ‍ ‍,1 Naomi Robson,2
Conventional teaching places emphasis Steven Bollipo,3,4 Global Online Alliance for Percutan Tech 2009;19:e32–5.
on the following landmarks for paracen- Liver Studies (GOAL), On behalf of the Global
7 McGibbon A, Chen GI, Peltekian KM, et al. An evidence-­
tesis: (i) in relation to the anterior supe- based manual for abdominal paracentesis. Dig Dis Sci
Online Alliance for Liver Studies (GOAL) 2007;52:3307–15.
rior iliac spine (ASIS): 5 cm superiorly and 1
Department of Gastroenterology, The Dudley Group 8 Thomsen TW, Shaffer RW, White B, et al. Paracentesis. N
medially;6 (ii) left or right lower quadrants: NHS Foundation Trust, Dudley, UK Engl J Med 2006;355:e21 https://www.​nejm.​org/​doi/​
2–3 cm lateral to the inferior rectus sheath 2
Biology Department, University of Toronto - full/1​ 0.​1056/n​ ejmvcm062234
border,7 (iii) midline approach within the Mississauga, Mississauga, Ontario, Canada
3
9 Sakai H, Sheer TA, Mendler MH, et al. Choosing the
linea alba: 2 cm below the umbilicus (this Department of Gastroenterology, John Hunter Hospital, location for non-­image guided abdominal paracentesis.
Newcastle, New South Wales, Australia Liver Int 2005;25:984–6.
is unconventional in the UK).8 However, 4
School of Medicine & Public Health, The University of
the use of absolute measurements is Newcastle, Callaghan, New South Wales, Australia
suboptimal as abdominal dimensions inev- Correspondence to Dr Keith Siau, Department of Guts UK 50 years old: onwards
itably vary (eg, in children, morbid obesity
or distortion due to chronic ascites). As
Gastroenterology, The Dudley Group NHS Foundation
Trust, Dudley, West Midlands, UK; ​keith@​siau.​org
and upwards
such, a relative approach for landmarks Twitter Keith Siau @drkeithsiau, Naomi Robson @
would seem more appropriate. robson_visuals and Steven Bollipo @stevenbollipo Guts UK is 50 years old. Despite many
We propose several alterations to the Collaborators Global Online Alliance for Liver Studies advances since its foundation, there
authors’ landmarks for paracentesis (GOAL) collaborators:Devika Kapuria, Atoosa Rabiee, is plenty left to be done. Funding for
(figure 1). We advocate the use of McBur- Gil Ben‐Yakov, Goutham Kumar, Stephen Congly, Juan research into gut, liver and pancreas
ney’s point as a rough surface landmark, Turnes, Renumathy Dhanasekaran, Rashid N Lui. diseases has always been modest
guided by abdominal percussion, with Contributors All authors contributed equally. compared with their clinical impact.1 The
the patient in supine position. This is Funding The authors have not declared a specific UK Medical Research Council was estab-
located one-­third of the distance between grant for this research from any funding agency in the lished in 19192 when gastroenterology
the right ASIS and the umbilicus, and is public, commercial or not-­for-­profit sectors. barely existed as a specialty—the British
traditionally used to localise appendicitis. Competing interests None declared. Society of Gastroenterology (BSG) was

Gut November 2021 Vol 70 No 11 2217