Haemorrhoids: modern diagnosis and treatment

ABSTRACT
Haemorrhoids present often to primary and secondary care, and haemorrhoidal procedures are
among the most common carried out. They may co-exist with more serious pathology, and correct
evaluation is important.In most cases a one-off colonoscopy in patients aged 50 or above with
flexible sigmoidoscopy in younger patients is reasonable. Many people with haemorrhoids do not
require treatment. Topical remedies provide no more than symptomatic relief—and even evidence
for this is poor. Bulk laxatives alone may improve symptoms of both bleeding and prolapse and
seem as effective as injection sclerotherapy. Rubber band ligation is effective in 75% of patients
in the short term, but does not treat prolapsed haemorrhoids or those with a significant external
component. Conventional haemorrhoidectomy remains the most effective treatment in the long
term, the main limitation being post-operative pain. Metronidazole, topical sphincter relaxants and
operative technique have all been shown to reduce pain. Stapled haemorrhoidectomy and
haemorrhoidal artery ligation techniques are probably less effective but less painful. Long-term
data are poor for all procedures, with many studies reporting only 1–3 years of follow-up data.
Haemorrhoids are common in pregnancy, occurring in 40% of women. They can usually be treated
conservatively during pregnancy, with any treatment delayed until after delivery. Acutely
strangulated haemorrhoids may be treated either conservatively or operatively. There is an
increased risk of anal stenosis after acute surgery, but the risks of sepsis and sphincter damage are
less significant than previously thought. The majority of patients who are treated conservatively
will still require definitive treatment at a later date.

INTRODUCTION
Lockhart-Mummery wrote on the diagnosis and treatment of internal piles 90 years ago in
the first edition of this publication. While a lot has changed since then, it is remarkable how much
of what he wrote remains true today. The potential pitfalls around diagnosis remain the same,
although far better tools have become available for investigation. The treatments that Lockhart-
Mummery espoused also remain in regular use with some modifications, but are no longer the only
options.
Haemorrhoids are common, reported incidentally at 40% of screening colonoscopies,2 and
interventions for haemorrhoids are carried out over 30 000 times each year in the UK.3
Haemorrhoids may also co-exist with more serious pathology, or be mis-diagnosed in its place.
Primary care prescriptions for topical haemorrhoidal treatments have been reported to increase in
the year before diagnosis with rectal cancer,4 so it is vital therefore that patients are investigated
appropriately before a diagnosis of haemorrhoids is made.
The current evidence base for the diagnosis and management of haemorrhoids is reviewed
here, reflecting on how the situation has changed over the last 90 years.

DIAGNOSIS
Lockhart-Mummery began his article by discussing the diagnosis of piles, and many still
use the terms ‘piles’ or haemorrhoids’ to refer to any perianal lesion or symptom. Most patients
who have symptomatic haemorrhoids (as surgeons understand the term) will present with bleeding
or prolapsing masses at the anal verge, but not all patients with these symptoms have
haemorrhoids.
Where the presenting complaint is of a prolapsing haemorrhoidal mass without any bleeding,
proctoscopy and clinical examination is sufficient to confirm the diagnosis. Altered rectal
bleeding, iron deficiency anaemia or other concerning symptoms such as change in bowel habit
should always prompt investigation for more proximal pathology. The uncertainty lies in deciding
how extensively to investigate a patient who presents with fresh painless post-defaecation rectal
bleeding consistent with haemorrhoids with no additional associated symptoms.
Flexible sigmoidoscopy for new onset fresh rectal bleeding will uncover either
adenocarcinoma or large polyps in just under 10% of patients,5 and should probably be the default
investigation. In patients over the age of 50 a colonoscopy is a more effective cancer screening
tool than flexible sigmoidoscopy. Where resources allow, this screening benefit may justify
colonoscopy in this age group even when symptoms are more suggestive of perianal pathology. In
younger patients the risk of cancer is lower and more selective use of colonoscopy is reasonable.

CLASSIFICATION
The traditional classification of haemorrhoids is into grades I–IV (or first to fourth degree),
where grade I haemorrhoids are purely internal, grade II haemorrhoids prolapse on defaecation but
reduce spontaneously, grade III need manual replacement, and grade IV cannot be replaced. This
grading system is frequently used as part of exclusion and inclusion criteria in clinical trials, but
in the authors’ view is of little use in clinical practice. While the distinction between internal and
prolapsing haemorrhoids has some relevance to treatment, progression from grade II to grade IV
is more an indication of gradually increasing haemorrhoidal size than of any change in underlying
pathology.
Probably of greater relevance is the presence or absence of an external component, as this
generally requires anaesthesia and formal excision if it is to be treated successfully; determination
of the patient’s wishes in this matter tends to determine office treatment versus operation.

MANAGEMENT
Many people with haemorrhoids require no treatment. In population-based studies, more
than half of those with haemorrhoids seen on clinical examination are entirely asymptomatic.
Around 20% of the population report rectal bleeding within the previous year, and only 40% of
those seeing rectal bleeding will ever seek medical advice for it.7 Haemorrhoids are very rarely
dangerous, and treatment is therefore carried out to improve quality of life. Patients with
symptomatic haemorrhoids do not as a group have decreased quality of life compared to the
general population, and so the ability of studies to demonstrate quality of life improvement after
treatment is limited.
A wide range of options is available for those who want treatment. There are many
comparative studies comparing treatment modalities, but they often involve only small numbers
of patients and short-term follow-up. Meta-analyses, where available, go some way towards
addressing the small patient numbers, but long-term follow-up data are still often absent. Selection
of treatment is therefore based only partly on the evidence and partly on personal experience and
expertise.

Outpatient management
The use of a bulk laxative may reduce haemorrhoidal symptoms of bleeding and itching in
symptomatic patients with or without pre-existing constipation, but the evidence is poor.8 Topical
remedies containing either local anaesthetic, or anaesthetic and steroid are frequently used for
symptomatic relief. There is no good evidence supporting their use, and they would not be
expected to have any long-term effect on haemorrhoids.
Interventions to treat haemorrhoids in outpatients aim primarily to disrupt the blood supply
into the haemorrhoid resulting in reduced vascularity and thereby a reduction in both haemorrhoid
size and per-anal bleeding. These techniques do not have a significant effect on any external
haemorrhoidal component, so where this is present and the patient wishes it dealt with, surgery is
the only option.
Lockhart-Mummery espoused injection sclerotherapy for the non-operative treatment of
haemorrhoids,1 and the technique remains the same although the sclerosant has changed. A
submucosal injection of 5% phenol in oil via a proctoscope causes sclerosis around the
haemorrhoidal vessels. This procedure remains in regular use despite limited benefit and reports
of occasional severe side effects,11 including intraprostatic injection. There is some short-term
benefit in terms of controlling rectal bleeding,12 but at 6 months after treatment the symptoms are
no better than if the patient had been treated with a bulk laxative alone.13 Results are particularly
poor with grade II haemorrhoids, with the majority of patients reporting worse symptoms at 3
years after treatment than before treatment.9 One situation where injection sclerotherapy may still
be considered is in the short-term management of bleeding grade I haemorrhoids for a patient who
does not wish or is not suitable for other treatment. Patients with these early haemorrhoids may
alternatively be treated with infra-red photo-coagulation with similar short-term results and fewer
reported side effects.
Rubber band ligation is currently the most popular outpatient treatment in the UK.11 It
uses a tight band to strangulate the haemorrhoid and its arterial inflow or, by application above the
haemorrhoid, to hitch the cushion back into the anal canal without destroying it.15 Results are
better than with injection, with 70–80% of patients responding in the short term, but it is less
effective than surgery, especially for larger prolapsed haemorrhoids.10 Significant bleeding occurs
after the procedure in 3% of patients, and is more likely in patients on antiplatelet agents or
anticoagulants, although the increased risk with aspirin is low.
About 1 in 20 patients may experience pain which may develop some hours after rubber
band ligation and which can take up to a week to subside and may disrupt intended meetings/ travel
plans. Accordingly, it is wise formally to consent individuals for this procedure.

Operative management
Haemorrhoidectomy was advocated by Lockhart-Mummery as the best method of treating
piles.1 However, many other surgical techniques have been developed since he wrote his article.
The driver for the development of new techniques has not been a lack of efficacy of conventional
haemorrhoidectomy but a desire to reduce post-operative pain.

Conventional haemorrhoidectomy
Haemorrhoidectomy involves excising both the internal and external component of each
haemorrhoid, while being careful not to damage the underlying internal sphincter, and ensuring
that adequate bridges of skin are left between the excised areas of anoderm to ensure there is no
circumferential scarring or subsequent anal stenosis. The procedure may be carried out with
scissors, diathermy or an energy device such as the LigaSure. The wounds are left open. The closed
operation, popular with those trained in North America, is a very different technique, allowing
preservation of more anoderm, thus permitting wound closure without stenosis. Outcomes are
similar. Post-operative urinary retention occurs in 5%, bleeding in 3%, and anal stenosis in less
than 1% of cases.
Medium-term results are excellent, with recurrence rates of under 2%. In the longer term,
recurrent prolapse has been reported in around 10% of patients after 7 years, although only small
numbers have been followed up for this length of time.
As an adjunct to conventional haemorrhoidectomy, a variety of techniques to reduce post-
operative pain are employed. It makes sense that the first post-operative stool is soft, so laxatives
should commence at least 2 days before surgery.25 A ligature to the haemorrhoidal pedicle might
impinge on the internal anal sphincter, thereby causing post-operative pain, so where possible
haemostasis should be achieved by diathermy alone.19 A dressing pack may increase pain and is
not necessary as long as adequate intra-operative haemostasis is achieved.26 27 A significant part
of post-operative pain may be caused by sphincter spasm, so some form of chemical sphincter
relaxation seems sensible, such as topical 2% diltiazem or topical 0.4% glyceryl trinitrate.
Post-operatively, the pain increases between days 3 and 5, and may partly be caused by
increasing bacterial olonisation of the wounds. Randomised trials with either systemic or topical
metronidazole showed a significant decrease in post-operative Pain.
Management of patient expectations contributes to the postoperative course. Patients
should be advised that the pain increase between days 3 and 5 is normal, that they will need 2
weeks off work, and that the wounds will take 6–8 weeks to heal. Early (but not too early) review
at 10 days avoids many first week calls and attendances and coincides with natural improvement.
There may be a modest reduction in early post-operative pain and a faster return to work if
a LigaSure energy device instead of diathermy is used for haemorrhoidectomy. However, longer
term results are equivalent and the increased cost with use of the disposable energy device has
damped enthusiasm.

Stapled haemorrhoidectomy/haemorrhoidopexy
When compared with conventional haemorrhoidectomy, postoperative pain after
haemorrhoidectomy/ haemorrhoidopexy is reduced and return to work is faster. Nifedipine may
be a useful treatment for those patients who develop pain and faecal urgency or tenesmus. Recto-
vaginal fistulas have been rarely reported. There are also anecdotes that the staple line may cause
penile injury during anal sex, so it is sensible to advise of this risk.
Individual trials have reported similar medium-term outcomes after either stapled or
conventional haemorrhoidectomy. Meta-analysis suggests that stapled haemorrhoidectomy is less
painful than conventional surgery but also less effective, with recurrence rates of 7% compared
with 2% in conventional surgery in the medium term.

Haemorrhoidal artery ligation

This group of operations involves ligating the blood vessels presumed to supply the
haemorrhoid. In the originally described technique, a Doppler probe was used to identify arterial
waveforms at six to eight circumferential points in the distal rectum. Each of these vessels was
then ligated with a deep suture placed per-anally. The technique has been frequently modified to
include a recto-anal repair, or mucopexy, in order to treat associated prolapse.
Reported results are very variable from series to series, with recurrence rates in the short
term ranging from 3% to 60%. At 3–4 years, the reported recurrence rate is between 14% and
25%.A systematic review showed an overall recurrence rate across 28 studies of 17.5%.43 The
technique is more effective in grade II and III haemorrhoids than in grade IV. The recurrence rate
at 5 years has been reported to be 28%.
The results appear to be slightly inferior to stapled haemorrhoidectomy in the short term,
but with reduced post-operative pain and analgesia requirements.43 A single study reporting at 48
months showed a worse result in the longer term, with 25% of patients experiencing recurrence
after artery ligation compared with 8% after stapled haemorrhoidectomy.
Early post-operative pain is also reduced compared with conventional haemorrhoidectomy.
Studies to date have not shown a difference in recurrence rate, but all had short follow-up times.
Results appear to be similar when the ligation is carried out without the benefit of a Doppler
probe, and this has the potential to reduce the cost of the procedure49 (while at the same time
undermining the already somewhat fragile raison d’etre of the original idea).

Special situations
Pregnancy
Some 40% of women will develop haemorrhoids during pregnancy, usually in the third trimester,
or within the first month after delivery.50 Symptoms are believed to resolve shortly after birth,
and intervention during pregnancy is therefore usually avoided.
Strangulated haemorrhoids
Haemorrhoids may strangulate when they prolapse and the tone of the internal anal sphincter
obstructs venous return. The resulting oedema may make reduction impossible, leading to
progressive strangulation. Traditionally, conservative management has been proposed for
haemorrhoids presenting acutely because of concerns over increased risks of anal stenosis,
sphincter damage, and sepsis after surgical intervention. But where patients are managed
conservatively, recurrent symptoms are likely in nearly 90% without subsequent treatment. Some
authors have argued for emergency surgery in almost all patients. Certainly the risk of portal
pyaemia in particular has probably been over-estimated, but anal stenosis rates after emergency
surgery remain high at 5–7%.

CONCLUSIONS
The majority of individuals with haemorrhoids are either asymptomatic or have symptoms that do
not cause them to seek medical advice, so they never receive treatment. Where people present with
rectal bleeding, it is important that other diagnoses are excluded.

Main messages
▸ Haemorrhoids should only be considered as the diagnosis once more serious causes of rectal
bleeding have been ruled out.
▸ Rubber band ligation is effective for small haemorrhoids but does not treat any external
component.
▸ Conventional haemorrhoidectomy offers the best long-term results.
▸ Newer procedures offer reduced perioperative pain and faster recovery, but with poorer results
in the long term.

Current research questions

▸ Do topical preparations provide better symptomatic relief than placebo?
▸ How do stapled haemorrhoidectomy and haemorrhoidal artery ligation compare in the long
term?
▸ What proportion of haemorrhoids which developed during pregnancy require subsequent
treatment?

Rubber band ligation is the most effective outpatient treatment but has only really been studied in
the short term and will not treat any associated external component. Some form of conventional
haemorrhoidectomy, whether open or closed, remains the most effective treatment for
haemorrhoids in the long term, and multiple strategies exist to reduce post-operative pain and
deliver short-stay surgery. Stapled haemorrhoidectomy and haemorrhoidal artery ligation are both
less painful than conventional
surgery, but are also less effective.