Keywords hemorrhoidal disease - hemorrhoids - collagen - morphometrics
Introduction
Hemorrhoidal disease (HD) has a high prevalence in the Western world, affecting about
one third of the population.[1 ] Despite the high prevalence and a large body of research, the etiology and pathogenesis
of HD has not yet been fully unraveled.
Hemorrhoids are normal vascular cushions filled with arterial-venous blood vessels
and are present in the submucosa of the anal canal. The expansion of the vascular
cushions is important in providing a watertight seal to the anus and plays a paramount
role in establishing fecal continence.[2 ] Hemorrhoids become a disease when the size of the hemorrhoidal complex enlarges
in a pathological way and may result in symptoms like pain, blood loss, and/or prolapse.[3 ] Hemorrhoids are mostly classified by the severity of prolapse according to the grading
system of Goligher et al.[4 ]
[5 ]
[6 ]
[7 ]
[8 ]
Until now, the exact pathophysiology of hemorrhoids remains unraveled, and several
theories are reported. The most widely accepted theory describes that the prolapse
of the hemorrhoidal plexus is initiated by a temporary return insufficiency of the
anal vascular system.[9 ]
[10 ] In case the return is hampered, progressive vascular dilation and hemorrhoidal hyperplasia
will occur, paving the way for hemorrhoidal prolapse.[11 ]
Besides, more and more evidence guides us towards the hypothesis that histological
alterations, that is, degeneration, of the hemorrhoidal plexus during aging play a
role in the development of HD.[1 ]
[12 ]
[13 ] The hemorrhoidal plexus is surrounded by a web of connective tissue, composed of
collagen, fibroelastic tissue, and muscle fibres.[14 ]
[15 ] The various fibers, sheets, and networks made of collagens are all extremely strong
and resistant to normal shearing and tearing forces.[16 ] Degeneration of these fibrous elements and alterations in collagen metabolism may
lead to destruction and loss of this supporting tissue.
Specific staining by Sirius Red (SR) is one of the most important stains to study
collagen networks in different tissues. Under polarized light, stained collagen bundles
appear green, red, or yellow. Polarizing light microscopy produces an image only of
material having repetitive, periodic macromolecular structure; features without this
structure are not seen. Since collagen fibers fulfil these requirements, they are
easily differentiated from the background, allowing for quantitative morphometric
analysis.[17 ] The young collagen fibers color yellow to green in comparison to the old fibers
that appear orange to red by polarized light.[18 ] The difference between these histological characteristics is based on the number
of crosslinks, which are abundant in mature collagen and lack in juvenile collagen.
Thin young green collagen can be interpreted as type III collagen, and thick red collagen
can be seen as old type I collagen. However, these interpretations should be made
with precaution since we speak of spectra of colors which can overlap between the
two types of collagens.[19 ] Therefore, in the present article, we chose to use the wording young collagen for
yellow/green collagen that lacks crosslinking and old collagen for orange/red collagen
rich in crosslinks.
In 2010, Willis et al. postulated that less connective tissue is associated with a
higher incidence of hemorrhoids.[13 ] They found that disturbances in collagen I/III and in collagen/protein ratios lead
to reduced connective tissue stability. Furthermore, loss of connective tissue may
result in dilation of supplying blood vessels, modifying the normal vascular anatomy.
Up to now, scarce reliable data exist on possible abnormalities in collagen composition
in HD. Therefore, the aim of the present morphometric study was to determine the quantity
and quality of collagen in patients with HD versus healthy controls, and to reconsider
the morphometrics of anal vascular structures in patients with HD.
Materials and Methods
Patients, donors, and recruitment
Patients with internal symptomatic HD grade III or IV who underwent standard hemorrhoidectomies
between January 2004 and June 2015 were included in the study group. During a hemorrhoidectomy,
an elliptical incision was made in the hemorrhoidal tissue extending proximally through
the dentate line to the upper limit of the hemorrhoid.[20 ] The surgeon excised the hemorrhoidal tissue including the mucosa and/or part of
the perianal skin.[21 ]
[22 ] Only patients > 18 years old with primary or recurrent HD were included. Patients
who had undergone Longo procedures or any other major anorectal surgery were excluded
from the study.[23 ] Further exclusion criteria were malignancies, aortic aneurysms, hernias, varicose
or other connective tissue diseases, patients on corticosteroid-, cytostatic- or radiotherapy,
and concomitant anorectal diseases (fistula, abscess, fissure, or polyps).
The resected hemorrhoidal specimens were fixed in 10% buffered formalin immediately
after the operation and were processed to embed in paraffin for further investigation.
The control group consisted of individuals who had donated their body to the department
of Anatomy and Embryology of Maastricht University and who had died of a natural death
between December 2016 and June 2018. Handwritten and signed codicils from the donors,
as required by the Dutch law, are kept at the Department of Anatomy and Embryology.
In the review of the past medical histories of these cadavers, symptomatic HD, as
well as the aforementioned exclusion criteria, were never listed as a condition they
had. In these cadavers, the pelvis was medially sawed through, leaving the halved
rectoanal area. After trimming, the emerged specimens were fixed in 10% buffered formalin
and processed to embed in paraffin.
The patients included in the control group gave premortem permission to use bodily
tissue in name of education and research. All patients included in the study group
gave their consent to undergo hemorrhoidectomy. Ethical clearance for the present
study and approval to use the resected specimens was obtained from the Maastricht
University Medical Centre ethical review board (file number 2017-0065).
Collagen Quantity and Quality
Collagen quantity was analyzed by collagen/total tissue ratio estimating the relative
amount of collagen. The quality of the collagen was analyzed by calculating the amount
of new collagen and the amount of old collagen.
The paraffin blocks were cut with a microtome (Leica 2245, Nussloch, Germany) and
the 5 µm thick sections were mounted on coated glass slides. After the paraffin was
removed, serial sections of each specimen were stained with hematoxylin and eosin
(HE) and Sirius Red (SR) according to the protocol of the local laboratory. For HE,
this entailed staining for 5 minutes with HE stains (Mayer Hematoxylin/Eosin protocol),
and for SR, the stain was applied by staining for 30 minutes with SR (0.1% of SR in
saturated aqueous picric acid).
Hematoxylin has a deep blue-purple color that attaches to nucleic acids and stains
nucleic blue. Eosin stains cytoplasm and extracellular matrix with varying degrees
of pink.[24 ] Sirius Red is a strong dye that identifies fibrillary collagen networks in tissue
sections.[25 ]
[26 ]
The SR-stained sections were observed with a microscope (Leica-DM4B, Nussloch, Germany)
with a 20-time objective. Pictures were made with normal and polarized light, creating
images of the same areas under the same conditions (exposure time of 17.8 milliseconds
versus 300 milliseconds).
Using a designed computer program (Leica Qwin, Cambridge, UK), the red stained collagen
was detected on the bright field image. Within this mask, the color of the birefringence
polarized collagen fibers was determined by the hue, saturation, and intensity settings
that determined the different color aspects of the birefringence collagen fibers.
The colors ranged from red, orange, yellow, and green to teal. The same parameters
defining threshold bands of hue, saturation, and brightness were applied to all the
images.[19 ]
Using polarized light, a distinction can be made between relatively new collagen,
which appears as green to yellow, and adult collagen, which appears as orange to red.
Collagen quantity is expressed as the percentage of collagen relative to the total
amount of tissue. The quality of the collagen is expressed as the percentage of new
collagen and the percentage of adult collagen relative to the total amount of tissue.[12 ]
Structural Vascular Alterations
Structural vascular alterations were estimated by the dilatation of the anal vessels
comparing HD patients with healthy controls. The perimeter of the anal vessel was
calculated using platelet-endothelial cell adhesion molecule 1 (CD31) endothelial
antibody staining. Platelet-endothelial cell adhesion molecule 1 is a transmembrane
glycoprotein with various functions in multiple physiologic and pathologic pathways,
and a very specific marker for endothelial cells.[27 ] The monoclonal anti-CD31 antibody recognizes a fixation-resistant epitope in endothelial
cells.[28 ]
Serial sections of each specimen were stained with CD31 endothelial antibody staining
(DAKO-M0823, dilution 1:100). After incubation, only the sections with distinct membranous
labelling for CD31 were taken into the analysis. Pictures of the sections with CD31
were uploaded on the computer and examined manually. The spaces surrounded by the
CD31 staining were considered vessels. Both arteries and venules were taken into account.
The demarcation of the vessel surface indicated by the CD31 stained endothelial cells
was noted and integrated in a computer algorithm for analysis. Anal vascular morphometrics
were specified into surface area, perimeter, and aspect of the vessel. The aspect
of the vessel is a ratio and is calculated by dividing the length of the vessel by
the width of the vessel. The surface area of the vessel is expressed in square micrometers
(micrometer2 ), the perimeter of the vessel is expressed in micrometers, and the aspect of the
vessel has no dimension (ratio).
Statistics
All statistical analyses were performed using IBM SPSS Statistics for Windows version
25.0 (IBM Corp., Armonk, NY, USA). Normal distribution was assumed after performing
the Shapiro-Wilk test. Multiple linear regression analysis was used to adjust for
baseline difference. Data were expressed as means ± standard deviation (SD). Correlations
between age and the percentage collagen were expressed as Pearson correlation coefficient
R . A p- value < 0.05 was considered statistically significant.
Results
Patients, Donors and Recruitment
The study group consisted of 22 patients who underwent a hemorrhoidectomy. The mean
age of the study group was 55.6 (31 to 78) years old, with a male to female ratio
of 15:7. The control group entailed 15 specimens with a significantly different mean
age of 82.7 (60 to 95) years old. The male to female ratio in the control group was
8:7.
[Table 1 ]
Table 1
A direct comparison of study versus control groups regarding collagen percentage,
with an asterisk indicating statistical significance
M:F ratio
Mean age in years (range)
Percentage of collagen of total tissue
p-value (percentage of collagen)
Percentage of young collagen
p-value (young collagen)
Percentage of old collagen
p-value (old collagen)
Study (n = 22)
15:7
55.6
(31–78)
62.1
± 13.8
0.0001*
0.00009
± 0.00008
0.001*
0.389
± 0.242
0.06
Control (n = 15)
8:7
82.7
(60–95)
18.7
± 14.5
0.0008
± 0.0008
0.588
± 0.286
Abbreviations: F, females; M, males.
Collagen Quantity and Quality
The mean percentage of total collagen was higher in patients with hemorrhoidal disease
(62.1 ± 13.8%) than in the control group (18.7 ± 14.5%; p = 0.0001). There was no difference in the mean percentage of collagen between males
and females (50.3 ± 24.4 versus 34.9 ± 25.7 micrometers; p = 0.08). Regression analysis showed that a higher age did not influence the percentage
of collagen (R = 0.86).
The percentage of young collagen was higher in the control group (0.0008 ± 0.0008%)
than in the study group (0.00009 ± 0.00008%; p = 0.001). The percentage of old collagen did not differ significantly between the
control group (0.588 ± 0.286%) and the study group (0.389 ± 0.242%; p = 0.06).
Fig. 1 (A) Sirius Red staining under normal light, indicating the relationship of young
collagen (yellow to green) versus old collagen (orange to red) in a healthy control
individual. (B) Sirius Red staining under polarized light, indicating the relationship
of young collagen (yellow to green) versus old collagen (orange to red) in a healthy
control individual'.
Fig. 2 (A) Sirius Red staining under normal light, indicating the relationship of young
collagen (yellow to green) versus old collagen (orange to red) in a patient with hemorrhoidal
disease. (B) Sirius Red staining under polarized light, indicating the relationship
of young collagen (yellow to green) versus old collagen (orange to red) in a patient
with hemorrhoidal disease.
Structural Vascular Alteration
A direct comparison showed a difference in the surface area of the vessels between
the study group (795.1 ± 1215.9 micrometer2 ) and the control group (1219.0 ± 1976.1 micrometer2 ; p = 0.003). The perimeter of the vessel was not different between the study group (131.7 ± 110.8
micrometers) and the control group (149.0 ± 117.3 micrometers; p = 0.12). The aspect of the vessel appeared larger in the study group compared with
the control group (3.10 ± 7.5 versus 2.16 ± 1.1, respectively; p = 0.02).
[Table 2 ]
Table 2
A direct comparison of study versus control groups regarding vessel data, with an
asterisk indicating statistical significance
Surface vessel in micrometer2
p-value (surface vessel)
Perimeter vessel in micrometer
p-value (perimeter vessel)
Aspect vessel
p-value (aspect vessel)
Study (n = 22)
795.1 ± 1215.9
0.003*
131.7
± 110.8
0.12
3.10
± 7.5
0.02*
Control (n = 15)
1219.0 ± 1976.1
149.0
± 117.3
2.16
± 1.1
Discussion and Conclusion
Discussion and Conclusion
The present study demonstrated that patients with hemorrhoids had increased total
collagen, decreased young collagen, and smaller surface area of the anal vessels compared
with healthy controls.
In other studies that examined the role of collagen in HD, an association was found
between reduced connective tissue stability and an increase in the incidence of HD.[12 ]
[13 ] However, the present study revealed a higher percentage of total collagen in patients
with HD, suggesting ample connective tissue stability in this group. The mean age
of the control group was significantly higher than that of the study group. The literature
reports that the total amount of collagen reduces when you get older.[29 ] However, the higher mean age in the control group cannot function as an explanation
for this outcome because the regression analysis showed that age does not influence
the percentage of collagen. Nonetheless, it has to be noted that the small sample
size of the present study could have affected the regression analysis. In the present
study, we found that patients with HD have an increased percentage of total collagen.
Multiple studies have reported elevated resting pressure in patients with hemorrhoids,
which could result in vascular stretching.[30 ] Mechanical vascular stretching has shown to stimulate collagen synthesis.[31 ] A possible hypothesis for the increased percentage of total collagen in HD could
be a result of the expanded vessel walls.
Furthermore, the present study showed that patients with HD have a lower percentage
of young collagen. This finding is not in line with previous research by both Willes
et al. and Nasseri et al., who demonstrated that patients with HD have a decreased
type I/III ratio. These studies made the direct interpretation of orange/red shades
being old collagen I and yellow/green tints as young collagen type III.[12 ]
[13 ] In the present study, we made the distinction between young and old collagen. Young
thin collagen fibers appear yellow to green while old thick fibers appear orange to
red.[18 ] A decreased percentage of young collagen implies a decreased amount of crosslinking
and, therefore, could lead to a reduced mechanical stability of connective tissue
in the hemorrhoidal complex. Nevertheless, the total amount of collagen was higher
in HD patients and a statistically significant difference does not imply that the
finding is clinically significant. Furthermore, a direct comparison of collagen percentages
cannot be made, since these studies only reported the ratio of the parameters and
not the separate percentages.
A closer look into alterations in anal vascular structures showed an increase in the
surface area and in the aspect of the vessel in healthy controls compared with patients
with HD. Remarkably, the comparative data on vessel perimeter does not appear to be
of significant difference between the two groups. It is postulated that the malfunctioning
hemorrhoidal complex in HD could lead to the distention of vessels. However, the distention
of the vessels was greater in healthy controls. This could be explained by the effect
of a lifelong erect posture among the latter.[32 ] Furthermore, it must be considered that the controls originate from postmortem bodies
that are spouted under pressure (between 0.2 and 0.3 Atm) with a formalin mixture
in the femoral artery. This process leads to a visible distension of the vessels.
One of the limitations of the present study is the relatively small number of patients
in both groups. Nevertheless, it is quite difficult to obtain specimens from hemorrhoidectomy
operations while taking the various inclusion and exclusion criteria into consideration.
The same accounts for the specimens from the control group, which has also shown to
be a difficulty in previous pathology studies.[12 ]
[13 ] Furthermore, the medical history of the control group does not mention HD, albeit
people can suffer from HD without reporting this to their medical doctor. Another
aspect of the present study that could diminish the power of the evidence is the difference
between the study group and the control group. The groups differ both significantly
in the male to female ratio and in the mean age. The amount and nature of collagen
are likely to be influenced by age. In a small study, this can possibly give rise
to bias. However, making use of a regression analysis can give insight in possible
confounders and help to correct for them.
The final limitation of the present study is that both arteries and venules were taken
into account when estimating the structural vascular alterations. The CD31 is a transmembrane
glycoprotein that is present in the endothelium of arteries and veins. As a result,
we were not able to make a distinction between these vessels. However, we were interested
in a mean increase of diameter of the vessels overall. Since we calculated the diameter
of arteries and veins in both groups, a possible difference between the groups will
be corrected.
The main strength of the present study lies in the fact that it can assist in unravelling
the true pathophysiology of HD by further considering the involvement of alterations
in collagen composition and in the anal vascular structure. Translating the findings
of the present study to the clinical practice can contribute to optimize the care
for patients with HD.
Although multiple options are available for the treatment of HD, the holy grail has
not been found yet and many patients with HD are still seeking for their ideal cure.
Therefore, it is important to have a better understanding of the pathophysiology of
HD and of the different mechanisms involved. The results of the present study suggest
that collagen can play a role in the formation of hemorrhoids. Still, the relationship
between collagen and downregulation of protein synthesis demands further exploration.
The reduced amount of young collagen in HD patients could be of interest for promising
mechanisms in HD treatments. For example, the use of hyaluronic acid to improve remodeling
of extracellular matrix is increasing in popularity.[33 ] Nevertheless, one must keep in mind that the disease underlying hemorrhoids is multifactorial
and that further research on this societal burden has to be performed.
