Introduction: Preoperative gastric ischemic conditioning (IC) improves the outcome of esophageal replacement gastroplasty and is associated with low morbidity. However, when the stomach cannot be used for esophageal replacement, a colonic replacement is required. The study aim was to assess the viability of right colon and terminal ileum IC in a rat model and the histological damage/recovery sequence and determine if neovascularization is a potential adaptive mechanism. Methods: The study was conducted in Rattus norvegicus with ileocolic vascular ligation. Seven groups of animals were established (6 rats per group) with groups defined by the date of their post-IC euthanasia (+1, +3, +6, +10, +15, and +21 days). Comparisons were made with a sham group. Viability of the model was defined as <10% of transmural necrosis. The evaluation of histological damage used the Chiu score in hematoxylin and eosin sections of paraffin-embedded specimens with CD31 immunohistochemical assessment of neovascularization by the median of submucosal vessel counts in 5 high-magnification fields. Results: Transmural colon necrosis occurred in 1/36 animals (2.78%) with no animal demonstrating transmural ileal necrosis. The maximum damage was observed in the colon on +1 day post-IC (average Chiu score 1.67, p = 0.015), whereas in the ileum, it was on days +1, +3, and +6 (average Chiu score 1.5, 1.3, and 1.17; p = 0.015, 0.002, and 0.015, respectively). In the +21-day group, histological recovery was complete in the colon in 4 (66.7%) of the 6 animals and in the ileum in 5 (83.3%) of 6 animals. There were no significant differences in quantitative neovascularization in any of the groups when compared with the sham group or when comparisons were made between groups. Conclusions: The tested animal model for IC of the colon and terminal ileum appeared to be feasible. Histological damage was maximal between the 1st and 3rd day following IC, but by day 21, recovery was complete in two-thirds of the rats. There was no evidence in this preliminary IC model that would suggest neovascularization as an adaptive mechanism.

Ischemic conditioning (IC) is the tissue adaptation resulting from partial or permanent suppression of vascularization. Preoperative ischemic conditioning (PIC), also known as the “delay phenomenon,” was developed by plastic surgeons to improve flap survival. This PIC technique mobilizes a flap but delays its transposition usually for 2–3 weeks, effectively achieving better adaptation to ischemia and diminishing the likelihood of necrosis.

In reconstruction after esophagectomy, the same principle applies requiring partial mobilization and devascularization of a segment of the digestive tract to construct a conduit; the stomach is the first option in adults, and when it is not available, the second option is the colon [1]. Esophagectomy has high morbidity (50%) and mortality (5–10%) [2], with ischemia of the conduit a principal determinant of significant complications, including anastomotic dehiscence, conduit necrosis, or stenosis. In a comparative study of gastric duct versus colonic duct, Briel et al. [3] reported that the global prevalence of duct ischemia was 9.2%, with no significant differences between the colonic and gastric ducts, whereas in the review by Gust et al. [1], the prevalence of ischemia of the colonic duct was between 0% and 10%. Although conducting a preoperative study of vascularization of the colon is not a systematic practice [1], an audit performed in the UK indicated that almost half of the esophagogastric surgeons surveyed consider performing preoperative angiography [4].

Different segments of the colon can be used to create the conduit, and a recent meta-analysis [5] stated that left coloplasty was the most frequent (70.9%) procedure, with a small difference in morbidity between left and right coloplasty (15.7% vs. 18.7%, respectively). In our team, we have more experience with right coloplasty [6, 7] which has the following advantages over left coloplasty: less disparity in the caliber between the ileum and esophagus, the possibility that the Bauhin valve acts as an anti-reflux mechanism, and the least variability in terms of vascularization. Arterial vascularization of the right colon is mainly determined by the ileocolic artery and the right branch of the middle colic artery, both communicated through the marginal artery (present in 99.8%, 95%, and 100% of cases, respectively), whereas the right colic artery has a less relevant role, a more variable origin, and is present as such in two-thirds of the cases [8].

Several techniques can improve conduit vascularity, such as PIC, supercharged microvascular augmentation, an omental patch, or exogenous agents (such as vascular endothelial growth factor) [9]. The use of gastric PIC was first analyzed in an experimental rat model by Urschel [10] in 1993. Later experimental models have shown that gastric PIC decreased the rate of dehiscence, improved oxygen delivery to the terminal end of the gastroplasty, and was associated with reduced collagen deposition [11]. Several recent clinical studies and meta-analyses show that PIC may decrease the rate of anastomotic dehiscence and the reoperation rate from leaks [2, 12], with low morbidity [2, 13].

The only available experimental animal study on the delay phenomenon in the colon we know of is by Souther and Vistnes [14] in which an ileocolic anastomosis was performed in rats after deliberate ligation of the penultimate and terminal branches of the ileocolic arcade. The study aim was to determine the viability of right colon and terminal ileum IC in a rat model, assess the histological damage/recovery sequence, and determine if neovascularization is a potential adaptive mechanism.

The study protocol was approved by the Ethics Committee for Animal Experimentation of the University of Barcelona, developed in the Bellvitge’s Animal Facility, and complies with the regulations of the European Union (directive 86/609 CEE) and May 1995 of the Generalitat de Catalunya regarding the protection of animals used for experimentation and other scientific purposes. The relevant ARRIVE 2.0 [15] and HARRP [16] guidelines were followed. Female Sprague Dawley rats weighing from 200 to 300 mg and aged from 8 to 12 months (Harlan U.K. Ltd., Shaw’s Farm, Blackthorn, Bicester, Oxon, OX OTP England) were used. Healthy rats were included after a quarantine period. The sample size was based upon a previous study conducted by our group on gastric PIC [17]. There were 6 study groups and a sham group (n = 6 each, 42 in total) grouped according to the number of days post-IC when they were euthanized: +1, +3, +6, +10, +15, and +21.

Ischemic Conditioning Protocol

Antibiotic prophylaxis was a preoperative dose of subcutaneous enrofloxacin (7.5 mg/kg) with buprenorphine analgesia (0.05 mg/kg). General anesthesia was isofluorane inhalation by using an induction dose at 5% and a 1–2.5% maintenance dose. A midline laparotomy was performed with identification, ligation (4-0 silk), and transection of the arterial and venous ileocolic vessels. The laparotomy wound was closed with a 3-0 silk suture, and the skin was closed with staples. The euthanasia method was by CO2 asphyxiation. To determine the optimal site for vessel ligation, 2 groups of 4 animals each were used: 1 group underwent tying off the main ileocolic vessel alongside the distal ileal branch (Fig. 1, white arrow), and 1 group underwent ligation of the ileocolic artery only (Fig. 1, green arrow). Two animals from the former group were euthanized due to deterioration of their clinical condition, with postmortem autopsy confirming cecal necrosis. Animals from the latter group were alive and well without exhibiting any clinical signs or symptoms on postoperative day 3. Therefore, ileocolic vessel ligation was chosen for all of the animals included in this study who were monitored daily until the postoperative day 3 and then every 48 h thereafter. Table 1 shows the pain assessment parameters. Euthanasia was performed if there was any evidence of animal distress. The following criteria for euthanasia were used: a ≥20% loss of weight, loss of consciousness by the animal, and/or an inability of the animal to move normally. During the first 3 days postoperatively, subcutaneous meloxicam 1 mg/kg every 24 h was administered for analgesia. All animals were fed and watered ad libitum.

Table 1.

Pain assessment parameters

 Pain assessment parameters
 Pain assessment parameters
Fig. 1.

Vascular anatomy of the ileocolic region in the rat. The white arrow indicates the initial ligation site that included the last ileal branch (necrosis of the cecum occurred in 2/4 animals). The green arrow indicates the definitive ligation point in the remaining animals. SMV, superior mesenteric vessels.

Fig. 1.

Vascular anatomy of the ileocolic region in the rat. The white arrow indicates the initial ligation site that included the last ileal branch (necrosis of the cecum occurred in 2/4 animals). The green arrow indicates the definitive ligation point in the remaining animals. SMV, superior mesenteric vessels.

Close modal

Outcome Measures

To determine the primary outcome measure (transmural necrosis), we accepted a colonic or ileal necrosis rate of <10% as our viability standard in accordance with a similar rate reported by Gust et al. [1] for eso-coloplasty. Secondary outcome measures included a histological damage score and quantification of neovascularization. The histological damage score has been described by Chiu et al. [18] for paraffin-embedded specimens stained with hematoxylin and eosin. Slides were categorized as 0: normal mucosa, 1: increase in the subepithelial space, 2: elevation of the epithelium, 3: moderate/severe intra/subepithelial edema, 4: transmucosal infarction, and 5: transmural infarction. Quantification of neovascularization was also performed on paraffin-embedded material immunohistochemically stained for the vessels with CD31 (ab28364; Abcam, Cambridge, UK). In each slide, the number of vessels in 5 high-magnification fields of both the colon and ileum at the submucosal level was counted, and the median number of vessels was compared with the reference (sham) group. Tissue samples were coded prior to histological analysis so that they were not identifiable by group. In addition to these groups, 8 more animals were used to define the ligation site of the vessels and handling of the samples. For histological analysis, the coded samples were analyzed by a surgeon supervised by a pathologist.

Statistical Analysis

Data were analyzed by using the Stata 13 program (StataCorp, College Station, TX, USA). The rate of necrosis was compared with the maximum expected proportion of 10% by using an exact binomial test. Differences in the rates of histological damage and vascular quantification were assessed by performing Fisher’s exact test and the Kruskal-Wallis test combined with a Fisher Pitman randomization, respectively. p values of <0.05 were considered to be indicative of statistical significance.

Histological Scoring

Among all animals (36), colonic transmural necrosis was evident in 1 (2.78%) of the +6-day group, with epithelial necrosis (Chiu score = 4) in 1 further rat. No ileal transmural necrosis was observed in any rat, and there was no need for a pre-emptive sacrifice to be made ahead of schedule. Figure 2 summarizes the Chiu score (minimum = 0, maximum = 30) in each group of animals. For the colon, maximal histological damage occurred in group +1-day with a significant difference when compared with the sham group (p = 0.015). In the successive groups, there was incrementally less quantitative histological damage in the other groups. There were 4/6 rats in the +21-day group that showed no histological damage (grade 0) and 2/6 animals in this group with only mild (grade 1) damage. Table 2 shows the differences in damage for both the colon and ileum between the 6 study groups and sham group. There were no significant differences in the colonic Chiu score for any other group. For the ileum, the maximum histological damage was evident in the +1-day group, with progressively less damage in the other groups. In the +21-day group, 5/6 animals had no histological damage (grade 0), and one showed mild (grade 1) damage only. There was a significant difference in histological damage between the +1-, +3-, and +6-day groups and sham group (Table 2).

Table 2.

Comparison of Chiu scores between the conditioned group and the sham group

 Comparison of Chiu scores between the conditioned group and the sham group
 Comparison of Chiu scores between the conditioned group and the sham group
Fig. 2.

Sum of the total Chiu score by group. Each column corresponds to the sum of the Chiu score of the animals belonging to that group.

Fig. 2.

Sum of the total Chiu score by group. Each column corresponds to the sum of the Chiu score of the animals belonging to that group.

Close modal

Blood Vessel Count

In each animal, the total number of vessels detected by CD31 immunohistochemistry in 5 high-magnification fields was determined. Animals with epithelial or transmural necrosis were excluded. The results are shown in Table 3 and graphically presented in Figures 3and4. In the sham group, the median number of ileal and colonic vessels was 13 (interquartile range [IQR] = 10–16) and 17.5 (IQR = 14.5–20.5), respectively. Overall, the median number of ileal and colonic vessels was 15 (8.5–21.5) and 15 (11–19), respectively. No significant differences were noted between the groups with IC and the sham groups or in any of the analyses between groups (p = 0.076). Overall, there was no difference in the mean weight between the animals with histological damage and those without (mean difference = 1.44 g, p = 0.42).

Table 3.

Median number of vessels* (and IQR) in the colon and ileum in euthanized animals after ischemic conditioning

 Median number of vessels* (and IQR) in the colon and ileum in euthanized animals after ischemic conditioning
 Median number of vessels* (and IQR) in the colon and ileum in euthanized animals after ischemic conditioning
Fig. 3.

Distribution of colon vessel count. Box plot of vessel count in the colon by post-IC euthanasia day.

Fig. 3.

Distribution of colon vessel count. Box plot of vessel count in the colon by post-IC euthanasia day.

Close modal
Fig. 4.

Distribution of ileum vessel count. Box plot of vessel count in the ileum by post-IC euthanasia day.

Fig. 4.

Distribution of ileum vessel count. Box plot of vessel count in the ileum by post-IC euthanasia day.

Close modal

This study showed that IC of the colon by ligation of the ileocolic vessels was feasible in this initial experimental model, with transmural necrosis of the colon observed in 1/36 animals (2.78%) and no transmural ileal necrosis. Importantly, ligation of the last ileal artery led to necrosis in half of the rats, a finding similar to that reported previously by Souther and Vistnes [14], so it must be preserved.

Overall, there was about a 10% risk of ischemia of the conduit (gastric or colonic) without PIC. Ischemic gastric conditioning provides a technically straightforward alternative with minimal morbidity [13], and colonic PIC could be a relatively simple strategy with similar benefits. Other more complex techniques can be used to improve perfusion (such as “supercharge”), but they have a high anastomotic leakage rate (42%) [19].

When designing an ileocolic conduit, knowing the anatomical variants is critical to preserve adequate vascularization. Colonic PIC enables prior direct assessment of the anatomy of the arterial supply to the colon, redirection of blood flow to the arteries for preservation, and prediction of vascular sufficiency before esophageal reconstruction surgery.

There is an ordered process of mucosal recovery following ischemia. Similar to gastric PIC [17], in our model, histological recovery was complete in two-thirds of the animals at 21 days, whereas the damage was mild in the remaining one-third, suggesting that 21 days would be adequate for performing an esophago-ileal anastomosis.

The mechanisms of IC have been studied mostly in the skin, whereas those in the digestive tract remain poorly understood. Furthermore, there is some confusion in the literature terminology, so we think that the most appropriate term to use is “preoperative IC,” and that “preconditioning,” “delay phenomenon,” or others should be avoided in this context.

Generally, these mechanisms can act on 3 levels: macrovascular, microvascular, and cellular effects. At the cutaneous level, an extensive study by Dhar and Taylor [20] suggested that after IC, there is a vasoconstriction period followed by permanent and irreversible dilation of the “choke vessels” that communicate with adjacent vascular territories (angiosomes), which is probably the most important mechanism of IC. At the intracellular level, there are modifications in gene expression, inhibition of apoptosis, nitric oxide-mediated vasodilatation, and metabolic reprogramming. At the microvascular level, there is neovascularization with anastomoses between the capillaries of the plasty and its recipient site [21]. Unlike gastric PIC [17], in our model, we were unable to demonstrate that neovascularization is an IC mechanism, which is consistent with Dhar and Taylor [20]. Although it was not a study objective, we did observe a considerable increase in the caliber of marginal vessels in some animals (Fig. 5), similar to what happens with right gastroepiploic artery in gastric conditioning [22].

Fig. 5.

Development of collateral circulation. The green arrow indicates the increase in diameter of the vascular arch of the ileum after ischemic conditioning.

Fig. 5.

Development of collateral circulation. The green arrow indicates the increase in diameter of the vascular arch of the ileum after ischemic conditioning.

Close modal

Our study had several limitations. Caution is needed when extrapolating the findings to humans since the rat has a relatively larger cecum in proportion to the rest of the intestine. The rat’s sex also may be important; Wu et al. [23] demonstrated that female mice tolerated ischemia-reperfusion-induced injury better than males. In our study, we used ex-breeding female rats to make better use of the available experimental animals given that when they would have finished their function and been euthanized anyway, along with being older, they would have been more similar to the adult human population that undergoes this type of procedure. Finally, there are currently more objective methods for quantifying neovascularization [24], but they are more expensive and require more resources.

Although this is a very early model of ileocolic PIC, we think it provides 2 relevant aspects for future studies using experimental animal models: histological recovery at 3 weeks is acceptable to perform an anastomosis, and preservation of the last ileal branch is key to avoiding necrosis of the cecum. Aspects to be analyzed in future studies could include the effect of PIC on blood flow in the colonic duct, anastomotic leak and conduit necrosis, dilation of the choke vessels as a mechanism of IC in the digestive tract, anatomical variations in the vascularization of the right colon in rats, effect of mechanical colonic preparation on PIC, and the role of experimental animal models in the surgical training of infrequent and complex procedures, such as this one.

Performing IC of the right colon and terminal ileum was feasible in this animal model, with almost complete histological recovery by ≤3 weeks. Neovascularization did not appear to act as an adaptive mechanism. More animal studies are required to assess the effect of ileocolic preoperative IC on conduit blood flow and anastomotic healing before considering its use in humans.

We thank Veronica Gomez for her support in preparing the surgical interventions and obtaining data. We also thank Alvaro Gimeno, Director of Animal House Bellvitge University of Barcelona, for the veterinary advice.

The study protocol was approved by the Ethics Committee for Animal Experimentation of the University of Barcelona at the February 12, 2012, meeting.

The authors have no conflicts of interest to declare.

The project was funded by grants from the Health Research Fund of the Carlos III Health Institute (FIS 98/0029-01 and IS 01/1260).

Rodriguez-Leon, De Oca, and Estremiana contributed to study conception and design. Rodriguez-Leon contributed to acquisition of data. Sabench, Jorba, and Rodriguez-Leon contributed to analysis and interpretation of data. Jorba, De Oca, and Rodriguez-Leon contributed to drafting of the manuscript. Miro, Bettonica, Aranda, and Farran contributed to critical revision.

All data generated or analyzed during this study are included in this article. Further enquiries can be directed to the corresponding author.

1.
Gust
L
,
Ouattara
M
,
Coosemans
W
,
Nafteux
P
,
Thomas
PA
,
D’Journo
XB
.
European perspective in Thoracic surgery-eso-coloplasty: when and how
.
J Thorac Dis
.
2016 Apr
;
8
(
S4
):
S387
98
.
2.
Michalinos
A
,
Antoniou
SA
,
Ntourakis
D
,
Schizas
D
,
Ekmektzoglou
K
,
Angouridis
A
,
.
Gastric ischemic preconditioning may reduce the incidence and severity of anastomotic leakage after οesophagectomy: a systematic review and meta-analysis
.
Dis Esophagus
.
2020 May
;
33
(
10
):
doaa010
.
3.
Briel
JW
,
Tamhankar
AP
,
Hagen
JA
,
DeMeester
SR
,
Johansson
J
,
Choustoulakis
E
,
.
Prevalence and risk factors for ischemia, leak, and stricture of esophageal anastomosis: gastric pull-up versus colon interposition
.
J Am Coll Surg
.
2004 Apr
;
198
(
4
):
536
41
.
4.
Fisher
RA
,
Griffiths
EA
,
Evison
F
,
Mason
RC
,
Zylstra
J
,
Davies
AR
,
.
A national audit of colonic interposition for esophageal replacement
.
Dis Esophagus
.
2017 May
;
30
(
5
):
1
10
.
5.
Brown
J
,
Lewis
WG
,
Foliaki
A
,
Clark
GWB
,
Blackshaw
GRJC
,
Chan
DSY
.
Colonic interposition after adult oesophagectomy: systematic review and meta-analysis of conduit choice and outcome
.
J Gastrointest Surg
.
2018 Jun
;
22
(
6
):
1104
11
.
6.
Farran-Teixidó
L
,
Miró-Martín
M
,
Biondo
S
,
Conde-Mouriño
R
,
Bettonica-Larrañaga
C
,
Danso
HA
,
.
Reconstrucción esofágica en un segundo tiempo: coloplastia y gastroplastia
.
Cir Esp
.
2008
;
83
(
5
):
242
6
.
7.
Farran Teixidor
L
,
Viñals Viñals
JM
,
Miró Martín
M
,
Higueras Suñé
C
,
Bettónica Larrañaga
C
,
Aranda Danso
H
,
.
Ileocoloplastia supercharged: una opción para reconstrucciones esofágicas complejas
.
Cir Esp
.
2011 Feb
;
89
(
2
):
87
93
.
8.
Bruzzi
M
,
M’harzi
L
,
Poghosyan
T
,
Ben Abdallah
I
,
Papadimitriou
A
,
Ragot
E
,
.
Arterial vascularization of the right colon with implications for surgery
.
Surg Radiol Anat
.
2020 Apr
;
42
(
4
):
429
35
.
9.
Enestvedt
CK
,
Thompson
SK
,
Chang
EY
,
Jobe
BA
.
Clinical review: healing in gastrointestinal anastomoses, part II
.
Microsurgery
.
2006
;
26
(
3
):
137
43
.
10.
Urschel
JD
.
Transplant delay will reduce anastomotic complications of gastric pull up for isolated esophageal atresia
.
Med Hypotheses
.
1993 Jun
;
40
(
6
):
379
80
.
11.
Reavis
KM
,
Chang
EY
,
Hunter
JG
,
Jobe
BA
,
Lineaweaver
WC
,
Schirmer
BD
,
.
Utilization of the delay phenomenon improves blood flow and reduces collagen deposition in esophagogastric anastomoses
.
Ann Surg
.
2005
;
241
(
5
):
736
47
.
12.
Heger
P
,
Blank
S
,
Diener
MK
,
Ulrich
A
,
Schmidt
T
,
Büchler
MW
,
.
Gastric preconditioning in advance of esophageal resection-systematic review and meta-analysis
.
J Gastrointest Surg
.
2017 Sep
;
21
(
9
):
1523
32
.
13.
Miró
M
,
Farran
L
,
Estremiana
F
,
Miquel
J
,
Escalante
E
,
Aranda
H
,
.
Puede el acondicionamiento gástrico disminuir la incidencia de dehiscencia anastomótica esofagogástrica cervical
.
Cirugía Española
.
2018 Feb
;
96
(
2
):
102
8
.
14.
Souther
SG
,
Vistnes
LM
.
The intestine and the delay phenomenon – a preliminary report
.
J Surg Res
.
1975 Jun
;
18
(
6
):
619
22
.
15.
Percie Du Sertid
N
,
Hurst
V
,
Ahluwaliaid
A
,
Alamid
S
,
Aveyid
MT
,
Baker
M
,
.
The ARRIVE guidelines 2.0: updated guidelines for reporting animal research
.
PLoS Biol
.
2020 Jul 14
;
18
(
7
):
e3000410
.
16.
Osborne
N
,
Avey
MT
,
Anestidou
L
,
Ritskes-Hoitinga
M
,
Griffin
G
.
Improving animal research reporting standards: HARRP, the first step of a unified approach by ICLAS to improve animal research reporting standards worldwide
.
EMBO Rep
.
2018 May
;
19
(
5
):
e46069
.
17.
Lamas
S
,
Azuara
D
,
de Oca
J
,
Sans
M
,
Farran
L
,
Alba
E
,
.
Time course of necrosis/apoptosis and neovascularization during experimental gastric conditioning
.
Dis Esophagus
.
2008 Jun
;
21
(
4
):
370
6
.
18.
Chiu
CJ
,
McArdle
AH
,
Brown
R
,
Scott
HJ
,
Gurd
FN
.
Intestinal mucosal lesion in low-flow states. I. A morphological, hemodynamic, and metabolic reappraisal
.
Arch Surg
.
1970 Oct
[cited 2018 Oct 27];
101
(
4
):
478
83
.
19.
Esmonde
N
,
Rodan
W
,
Haisley
KR
,
Joslyn
N
,
Carboy
J
,
Hunter
JG
,
.
Treatment protocol for secondary esophageal reconstruction using “supercharged” colon interposition flaps
.
Dis Esophagus
.
2020 Oct
;
33
(
10
):
doaa008
.
20.
Dhar
SC
,
Taylor
GI
.
The delay phenomenon: the story unfolds
.
Plast Reconstr Surg
.
1999
;
104
(
7
):
2079
91
.
21.
Schreiber
T
,
Salhöfer
L
,
Quinting
T
,
Fandrey
J
.
Things get broken: the hypoxia – inducible factor prolyl hydroxylases in ischemic heart disease
.
Basic Res Cardiol
.
2019
;
114
(
3
):
16
.
22.
Akiyama
S
,
Kodera
Y
,
Sekiguchi
H
,
Kasai
Y
,
Kondo
K
,
Ito
K
,
.
Preoperative embolization therapy for esophageal operation
.
J Surg Oncol
.
1998 Dec
[cited 2019 Mar 23];
69
(
4
):
219
23
.
23.
Wu
M
,
Rowe
JM
,
Fleming
SD
.
Eicosanoid production varies by sex in mesenteric ischemia reperfusion injury
.
Clin Immunol
.
2020 Nov
;
220
:
108596
.
24.
Vermeulen
PB
,
Gasparini
G
,
Fox
SB
,
Colpaert
C
,
Marson
LP
,
Gion
M
,
.
Second international consensus on the methodology and criteria of evaluation of angiogenesis quantification in solid human tumours
.
Eur J Cancer
.
2002
;
38
(
12
):
1564
79
.

Additional information

Meeting presentation: 25th Congress of the Spanish Society for Surgical Research, Barcelona, December 2019. Abstract: BJS 2020;107 (S1):5–20.