Healing and pain after haemorrhoidectomy

808 BRIT. J. SURG., 1964, Vol. 51, No. 11, NOVEMBER

The neatness and precision which marked his operating was a characteristic of his whole life, for he was the most methodical and tidy of men and his life was organized and disciplined in every detail. His papers and correspondence were always in perfect order and meticulously filed and indexed, and his mind was similarly ordered so that he seldom forgot any fact which he had read or learned and always seemed able to recall it whenever he needed it.

He was very much a creature of habit and some of these habits continued long after they had lost the purpose which brought them into being and this was a source of some amusement to both staff and students alike. Thus, the operating lamp which he is shown wearing in the photograph became a sort of trade- mark and always featured in the student Christmas show. It was given to him originally by de Martel, in Paris, and all his operating caps were made to accommodate it. He continued to wear it in spite of all advances in theatre lamps, oblivious of the fact that it had become so ineffective that he usually had to ask his assistant whether it was on or not !

Apart from surgery, Lambert had a life-long interest in the sea and in ships. He served in the Royal Navy in two world wars and the tradition of the Navy and of a ship’s company coloured the whole of his life. He often used to say of the Unit, ‘There can only be one captain of a ship, but every member of the crew must feel that they are important and

matter.’ He was a strong disciplinarian, and woe betide the student or junior member of the staff who sought to take liberties with him. At the same time he was the most loyal of colleagues, and every member of his staff knew that the Professor would stand by them through thick and thin, and would always have their best interests at heart.

I t was his sense of fairness and loyalty which made him so beloved of students and staff alike, and the real affection in which he was held was shown by the way his old students kept in touch with him and always came to see him when they returned to Cardiff, and he in his turn kept in touch with many of his old students all over the world by regular correspondence. I t was demonstrated, too, by the fact that the students in the Medical School, at the time of his death, spontaneously subscribed to com- mission the painting of a portrait of the Professor, which now hangs outside the Elize Nixon ward, where he imparted to so many generations of students, not only something of his profound surgical know- ledge, but also the ideals which inspired his own practice of surgery.

Not a few of those students now hold positions of responsibility and influence in many parts of the world, and no doubt the inspiration and example of their beloved teacher, Lambert Rogers, live on in them and are passed on in turn to those whom they teach. A. S. A.

ORIGINAL PAPERS

HEALING AND PAIN AFTER HAEMORRHOIDECTOMY BY J. McK. WATTS, R. C. BENNETT, H. L. DUTHIE, AND J. C. GOLIGHER

FROM THE UNIVERSITY DEPARTMENT OF SURGERY, THE GENERAL INFIRMARY AT LEEDS

hlosT previous observations on healing after haemorrhoidectomy seem to have been based on the behaviour of the external wounds, supplemented by palpation of the anal canal at varying intervals after operation. Actually these methods provide little reliable information regarding the healing of the wounds in the anal canal itself, display of which requires the passage of an anal speculum or proctoscope. lh i s manceuvre has usually been avoided during the phase of healing because of the discomfort involved and the necessity for a general anaesthetic. Recently we have become interested in the relative merits of different forms of haemorrhoidectomy and thought that it would be instructive, as part of the plan of evaluation of these various operations, to attempt endoscopic examination of the anal canal in the early postoperative period. In this paper we describe the results of our inquiry, in which the observations on the intra-anal wounds are correlated as far as possible with the technique of operation, the rate of healing, and the incidence of later postoperative complications.

-

CLINICAL lMATERIAL AND METHODS OF INVESTIGATION

I . Selection of Operation.-The series include 104 patients (67 men and 37 women), submitted to various forms of haemorrhoidectomy in the Professorial Surgical Unit, Leeds, between March, 1962, and April, 1963.

The operations performed were :- Excision with high ligation (Salmon, quoted

by Allingham and Allingham, 1896) 5 patients Excision with low ligation (Miles, 1919;

Milligan, Morgan, Jones, and Officer, 1937) 38 patients

Excision with primary suture (Mitchell, 1903 ; Ferguson and Heaton, 1959) 16 patients

Submucosal excision (Parks, 1956) 29 patients Excision with clamp and cautery (Cusack,

1846; Farquharson, 1962) 16 patients

With rare exceptions the operations were performed by the four authors themselves. They represent most of the haemorrhoidectomies done in this Unit during the period mentioned, but a few patients with

WATTS E T AL. : HAEMORRHOIDECTOMY 809

associated conditions, such as fissure or fistula, were not included. Some words of explanation are required on the method of selection of operation for individual cases. No organized system of random allocation was practised. Instead, each type of operation was taken in turn and accepted as the routine procedure, until sufficient cases had been performed to demonstrate the vagaries of wound healing after that particular technique. During this time virtually all the haemorrhoidectomies performed were that type, but a few patients, who should have had a submucosal operation, were treated instead by a standard low ligature and excision because they happened to come late on a long operating list and a quicker method was preferred. On two other occasions the submucosal technique was not used because excessive haemorrhage was feared from exceptionally large and vascular piles.

2. Assessment of Postoperative Pains.-Each patient was interviewed and examined before a panel of six or seven members of the Unit surgical staff on the eighth postoperative day. Initially he was questioned regarding the severity of the discomfort experienced during the first z days and nights after operation, special note being taken of how well he slept, the amount of analgesic required, and the reports of the nursing staff and house-surgeon. Inquiry was then made about the severity of pain caused by the first and subsequent bowel actions after operation. A careful digital examination of the anal region was then performed by a member of the panel and a report given of the amount of discomfort and spasm elicited by this procedure. Finally, with the aid of all this information, members of the panel were asked to write down their individual opinion of the severity of the pain using one of the following five categories : (A) Almost pain-free; (B) Slight; ( C ) Average; (D) More than average; (E) Severe. The majority opinion was accepted as the final verdict, it being notable, incidentally, that there was usually fairly close agreement between the opinions of individual members of the panel.

For the purposes of statistical evaluation of the results, the following numerical values have been given to each category of pain. A= I, B=z, C = 3 , D-4, E-5, so that the greater the degree of postoperative pain in any patient the greater the score.

3 . Examination of Zntra-anal Wounds.-The wounds were re-examined 10 days after operation with the aid of a brief general anaesthetic. The condition of the external wounds and intervening skin bridges was noted, the anal canal was then palpated digitally, and finally a careful inspection of the wall of the anal canal was conducted using a bivalve speculum. With the aid of this instrument, it was possible to examine completely each of the haemorrhoidectomy wounds and the intervening skin bridges. The dimensions and shape of the wounds and the intervening skin and mucosal bridges were then recorded diagrammatically. It was found necessary to conduct the examination with minimal opening of the blades of the speculum to avoid artificial separation of the wound edges. Additional points recorded were the presence of sloughing and suppuration, and whether the main pile ligatures had separated or were still in situ.

4. The state of healing of the anal wounds was reviewed at the Out-patient Rectal Clinic 6’ weeks from the date of operation. At this examination the anal canal was again palpated and inspected with the aid of a proctoscope.

5 . A further out-patient interrogation and examination was carried out 6 months after operation. The patient was questioned and examined fully in order to determine the presence of skin tags, fibrosis, stricture, or recurrent haemorrhoids.

FIG. I.-Excision with High Ligation. View into anal canal at completion of operation.

Lower figures: Diagrammatic representation of opened-out anal canal to illustrate: ( I ) State of wounds at completion of operation, with ligatures applied at level of anorectal ring; (2) State at IO days, showing extensive intra-anal wounds and breakdown of mucoszl bridges at the anorectal ring. (a-r.r., anorectal ring; p.l., pectinate line; a.v., anal verge.)

FINDINGS: STATE OF WOUNDS AT 10 DAYS

After Excision with High Ligation.- Technique.-This method, introduced by Frederick

Salmon, was the first form of ligature and excision operation for piles. Separation of the pile was begun by a cut at the mucocutaneous junction, which was then carried up on either side of the pile. The pile, with its overlying mucosa, was stripped off the internal sphincter muscle as far as the anorectal ring, where it was tied off and the excess of tissue excised. The disadvantage of this operation was said to be that it left extensive raw areas in the wall of the anal canal, which, on healing, were liable to produce much fibrosis and frequent strictures (Anderson, 1909; Milligan, 1929). Consequently it fell out of favour and has been replaced by the low ligature operation.

In the 5 patients treated by this method the technique was modified in two respects: some of the

X I 0 BRIT. J. SURG., 1964, Vol. 51, No. 11, NOVEMBER

FIG. 2.-Excisio11 with Low Li,yarion. A, Section through wall of anal canal. The pile has been drawn down and ligatured at the lower edge of the internal sphincter; it is believed to he held at this level by inclusion of the longitudinal fibres in the pedicle. B, View into anal canal at completion of operation illustrating the site of the ligature.

Lower figures: Diagrammatic representation of opened-out anal canal to illustrate: (I! Level of ligatures at completion of operation; 2) Expected findings at 10 days, with preservation of intra-anal

mucosa; (3) Actual findings at 10 days. Extensive intra-anal wounds extending up to level of anorectal ring; (4) In one-third of patients breakdown of one or more mucosal bridges between adjacent wounds.

anal canal and perianal skin relative to each pile was excised, and the operation was performed either with the aid of a bivalve speculum, or after vigorous stretching of the anal canal.

Findings at 10 Days in 4 Patients (I patient declined examination).-

I. Wounds (Fig. I) : As expected, extensive wounds were present in ail cases, extending from outside the anal verge up to the level of the ligatures at the anorectal ring. The wounds were ;-I in. in width throughout their length, but slightly wider in the vicinity of the ligatures in the upper anal canal. Three of the 4 patients had slough covering the intra-anal wounds.

2. Skin and mucosal bridges (Fig. I ) : One mucosal bridge was broken in 3 of the 4 patients, the site of breakdown in all instances being high in the canal at the level of the ligatures. Oedematous skin tags were not present in any of the 4 cases.

After Excision with Low Ligation- Technique.-To avoid the extensive granulating

areas in the anal canal associated with the stripping and high ligation of Salmon’s operation, Miles (1919) evolved the method of haemorrhoidectomy with low ligation. In this technique the initial cut is placed, not at the mucocutaneous junction, but lower down in the perianal skin, and is carried up on either side of the pile to a point immediately above the dentate line. The pile is then freed from the underlying tissues up to the level of the lower border of the internal sphincter, and a ligature is applied, into which as much of the mucosal-covered part of the pile as possible is drawn. It was claimed that this ligature, when tied, held the mucosa down in the lower part of the anal canal. Milligan and others (1937) have emphasized the advantage of this method in preventing retraction of the pile pedicles, and therefore maintaining mucosal coverage of the anal canal during healing. Their concept of the state of affairs at the conclusion of operation and during subsequent healing might be depicted as in Fig. 2.

In our series of haemorrhoidectomies with low ligation, we have followed the Milligan-Morgan technique precisely, except that in 19 of the 35 patients we have immediately preceded operation by a gradual, but forcible, dilatation of the anal canal with four fingers for a timed 4-minute period.

Findings at 10 Days in 35 Patients (3 patients declined examination).-

I. Wounds (Fig. 2): There was little evidence of healing in any of the 105 wounds. In all cases the raw areas extended, not, as expected, to just above the dentate line, but up to and sometimes beyond the anorectal ring. The wounds were broad, both externally and in the mucosal-covered part of the anal canal, but usually narrowed at the level of the dentate line, giving them a dumb-bell shape. The size of the external wound corresponded to the amount of skin sacrificed at operation, and only 9 of 135 external wounds were closed at the time of examination. At the level of the dentate line the wounds usually narrowed to approximately +in., and 34 of them were actually closed at this point. Above the dentate line the wounds became more extensive, being usually &I in. in width. Only 3 of the wounds were closed in this situation. The ligatures had separated in most cases but slough remained in more

WATTS ET AL. : HAEMORRHOIDECTOMY 811

than half of the wounds. Free pus was present in 7 patients.

2. Skin and mucosal bridges (Fig. 2 ) : In the performance of this operation great care was always taken to preserve adequate mucocutaneous bridges up to the level of the ligatures. I t is therefore surprising that in 11 of the 35 patients there was partial or complete loss of one or more bridges between the wounds, and in 2 of these 11 patients two mucosal bridges had gone, resulting in a considerable area of granulating wound. The site of breakdown of the bridges in 10 of these patients was high in the anal canal, often extending down to the dentate line. One patient, however, had lost an external skin bridge, but this may have reflected a wide excision of external skin at operation. Oedematous skin bridges, causing painful postoperative external piles, were observed in only 3 of the 35 patients.

Preliminary stretching of the anal sphincters made no apparent difference to the size of the wounds, or to the incidence of intervening bridge breakdown.

After Excision wi th P r i m a r y Suture.- Tech&pe.-In 1903 Mitchell described a method

of haemorrhoidectomy in which the pile was clamped radically across its base, and the tissue projecting beyond the clamp excised. A running catgut stitch was inserted loosely over the clamp, until the entire cross-section of the pedicle had been transfixed. The clamp was then released and slipped out of the embrace of the suture, which, at the same moment, was tightened and tied. In this way primary apposi- tion of the haemorrhoidectomy wound was achieved, and if, as was Mitchell’s practice, the skin-covered part of the pile was also included in the clamp, no raw area at all was left at the completion of the operation.

More recently haemorrhoidectomy with primary closure by a somewhat different technique has been advocated by Ferguson and Heaton (1959). In the first instance they carry out a complete excision of the pile, including its skin-covered component below and extending superiorly to the anorectal ring. T o facilitate this removal they employ a speculum, such as Sim’s vaginal instrument. The wound in the lining of the anal canal, and in the perianal region, is then closed by one continuous fine catgut stitch.

The haemorrhoidectomies with primary suture in our series were performed using either of these methods, 7 by the Mitchell and 9 by the Ferguson- Heaton technique. In all cases the operation was preceded by stretching of the anal canal for 4 minutes, so that the piles could be easily delivered through the anus, making the use of a speculum unnecessary.

Findings at 10 Days in 16 Patients.- I. Wounds (Fig. 3): Partial or complete separation

at the suture Iine was the usual finding, especially in the upper anal canal where the wounds were usually $-;in. in width. Thus, of the total of 47 wounds in 16 patients, 25 remained closed in their external part, 16 at the level of the dentate line and only 8 above this point.

2. Skin and mucosal bridges (Fig. 3): The findings were similar to those found after the low ligation technique. One or more mucosal bridges had sloughed in 5 of the 16 patients. In three instances this loss of bridge extended beyond the dentate line into the cutaneous compartment of the anal canal

as far as the anal verge. Two patients had lost two mucosal bridges with resulting fusion of adjacent wounds, but no case had completely circumferential raw areas. Oedematous skin tags were present in 3 of the 16 patients at the time of review.

FIG. 3.-Excision with Priniary Suture. View into anal canal at completion of operation.

Lower figures: Diagram of opened-out anal canal to illustrate: ( I ) State of wounds a t completion of operation; (2) State at 10 days, revealing separation of wounds and breakdown of mucosal bridges in one-third of patients.

Use of either the Mitchell or the Ferguson-Heaton technique did not make any obvious difference to the results.

After Submucosal Haemorrhoidectomy.- Technique.-As its name implies, this operation,

introduced by Parks in 1956, involves removal of the vascular and connective-tissue elements of the pile by dissecting them out from the overlying mucosa. This is accomplished by exposure of the pile throughout its length, with the aid of a bivalve anal speculum. The submucous and subcutaneous tissues are infiltrated with a weak solution of adrenaline for haemostatic purposes, and a longitudinal, inverted, racket-shaped incision is made in the covering of the pile, the ‘handle’ being placed in the mucosa and the rounded portion in the skin of the anal canal and perianal region. Mucocutaneous flaps are then raised on either side, and, starting below, the pile is dissected off the underlying sphincter muscles to the upper end of the anal canal. The pedicle thus defined is tied off with fine catgut, and the rest of the pile excised. The mucosal flaps then fall back on to the raw area, covering most of it, except externally, where a small open wound remains. One or two catgut stitches may be inserted to bring the flaps


together and to fix them to the underlying internal sphincter .

We have used either the technique outlined above .-a- a slight modification of it, in which the handle of

FIG. 4.-.Subm~icosal Excirion. Sketches of operative technique : A, B, Line of preliminary mucosal incision) C , Dissection of hood of mucosa off the haemorrhoid; D, Skin incision externally; E, Deep dissection, stripping haemorrhoid off internal sphincter; P, HiFh ligation of haemorrhoid at level of anorectal ring, under hood of mucous membrane G, Mucosa sutured to lower end of internal sphincter.

Lozuer Aqures: Diaarams of anal canal opened out to illustrate: I: State of wounds at completion of operation; ( 2 ) Minimal

rntra-anal mound\ at 10 davs

the incision is omitted and the dissection is accomplished entirely through the lower circular incision, the mucosa being separated from the pile as a cowl or hood (Fig. 4). Of the 27 operations, 10 were performed using the former method, 17 by the latter. A routine preliminary stretching of thc anal sphincter was not performed, but we believe that the presence of a widely opened bivalve speculum in the anal canal for the 30-40 minutes required for this operation probably has as much dilating effect on the sphincter as a formal sphincter-stretch.

Findings at 10 Days in 27 Patients (2 patients were not examined).-

I . Wounds (Fig. 4): These will be considerd in two groups according to the technique used :-

a. With Longitudinal Division of Mucosa.-In all cases linear wounds were present, extending from outside the anal orifice to the anorectal ring. Healing was never complete, but the wounds were rarely greater than $ in. in width.

b. Without Mucosal Incision.-It was expected to find an intact anal canal mucosa in this group, but this was present in relation to only half of the u-ounds. In the others, the wound extended upwards for a variable distance towards the anorectal ring, in- dicating that in some way the mucosal hood had undergone partial splitting or sloughing subsequent to operation. On several occasions the submucous catgut ligature on the pile pediclc had ulcerated through the undermined mucosa to produce a sloughing mucosal defect high in the anal canal.

2 . Skin and mucosal bridges: The mucosal bridges were all wide and intact in their entire length. Oedematous skin tags were present in 8 of the 27 patients.

After Excision with Clamp and Cautery.- Technique.-This form of haemorrhoidectomy

enjoys little popularity at the present time, but there are a few surgeons, such as Farquharson (1962), who have used it regularly for years. We have followed exactly his technique for its performance.

As a preliminary, the anal sphincters are stretched as described in connexion with other forms of haemorrhoidectomy. The skin and mucosal-covered part of the pile are then grasped with artery forceps and drawn down to display its maximum extent. A Farquharson distal-jointed clamp, with plates of plastic material on the under-aspect of the blades, is then applied radially across the base, the entire pile, including its external skin component, being included. Dry gauze packs are then wrapped underneath the plastic plates and handles of the clamp in order to insulate the skin from the heat of the cautery, and all but a short fringe of the pile tissue projecting bevond the blades of the clamp is excised with scissors. The cauterizing iron (which may be a soldering iron, or a special rectangular metal block on a suitable handle) heated to a temperature just short of red heat, is then applied firmly to the remnant of the pile. The pressure is maintained for at least 2 or 3 minutes to destroy the projecting tissue and also to heat the metal of the blades, thereby cauterizing and sealing together the part of the pile gripped by the clamp. At the end of this time the screw of the clamp is loosened; then, with a sudden movement, the blades are separated,


allowing the crushed, cauterized fringe of skin and mucosa to slip out of their grasp. Occasionally, on release of the clamp, the edges of the wound separate and bleed, requiring a touch of diathermy or a stitch of catgut, but usually they remain united in the linear eschar.

Findings at 10 Days in 15 Patients (I patient declined examination).-

I . Wounds (Fig. 5 ) : The striking feature about the wounds after the clamp and cautery operation was the great amount of sloughing present, and the extensiveness of the raw areas. Residual necrotic tissue was noted in g of the 15 patients, and the wounds in all 15 were widely separated throughout their length, being I in. or more at the dentate line and usually narrowing slightly above this level.

2. Skin and rnucosal bridges (Fig. 5 ) : So extensive was the mucosal damage following this operation that 10 of the 15 patients had destruction of one or more mucocutaneous bridges. Two patients had two mucosal bridges broken down, and 3 patients had complete circumferential destruction of the mucosa in the anal canal. The commonest site for the bridges to have sloughed was at the level of the dentate line. Oedematous skin tags were present in only I of the 15 patients.

SEVERITY OF POSTOPERATIVE PAIN The severity of postoperative pain, as judged by the

criteria already described (see above), is indicated in Table I . These findings will be discussed in a later section.

REVIEW AT 6 WEEKS One hundred patients attended for review in the

out-patient department 5-6 weeks after operation. Most of the wounds were found to be entirely healed but, in 15 patients, one or more of the external wounds were partly open. The number of patients with unhealed wounds after the different operations is shown in Table If.

REVIEW AT 6 MONTHS Of the 104 patients originally submitted to haemor-

rhoidectomy, I could not be induced to reattend, I had gone abroad, and 2 could not be traced because of change of address. One hundred patients were therefore followed and re-examined 6 months after operation. A thorough rectal examination and interrogation at this stage elicited the following facts :-

Wound Healing.-None of the patients had unhealed wounds at this stage.

Anal Fibrosis and Stenosis.-We have attempted to gauge the amount of anal fibrosis or stenosis in these patients by recognizing three grades of severity: (I) Stenosis preventing the insertion of the index jinger: None of our IOO patients fell into this category, for digital examination was possible in all of them. (2) Stenosis allowing digital examination, but preventing the passage of a proctoscope 3 in. in diameter : Four of our cases had this amount of narrowing; 2 of these were symptomless, but the other z had slight difficulty with defaecation attributable to the stenosis. The latter 2 have been treated by regular passage of a dilator for a few weeks, with complete relief of their symptoms. ( 3 ) Easily palpable fibrosis,

but digital and proctoscopic examination possible without dtficulty : Eight of our patients satisfied this criterion; none of them have any symptoms referable to the induration.

FIG. 5.-Clanip and Cautery Excision. Sketches of operative technique. -4, Haemorrhoid drawn down prior to application of clamp; B, Clamp applied radially across haemorrhoid and excess tissue excised; C, Cautery applied after shielding with gauze; D, Appearance of anus at completion of operation.

Lower jigtires: Diagrams of anal canal opened out to illustrate: ( I ) Linear eschars at completion of operation; (2) Broad extensive granulating wounds throughout anal canal at 10 days; (3) Break- down of mucosal bridges between adjacent wounds at 10 days; (4) Circumferential mucosal breakdown in one-fifth of patients at 10 days.


Table Z.-SEVERITY OF POSTOPERATIVE PAIN -.

I OPERATION AND NUMBER OF CASES

Suhmucosal Excision with Excision Low Ligation 1 (28) (35)

Unhealed wounds 1 3 ( 1 1 per cent) 3 (9 per cent)

SEVbKITV OF PAIN Excision with Low Ligation Submucosal Excision with Excision with

Score Excision With Sphincter Stretch Without Sphincter Stretch Primary Suture Clamp and Cautery --___I_- __ _ _ _ - - - ~ - ~

.4 (almost none) 4 0 0 3 5 4 2 4 3 5 6 5 6 4 3 2 I 4 3

C (average) 3

E (severe) 5 1

R (less than average) z

1) (more than average) 4

Excision with Excision with Excision with Primary Suture Clamp and Cautery High Ligation

(16) (16) ( 5 ) -___ -____ 2 (12 per cent) 3 (19 per cent) 4

Total number of cases I 29 I I9 I I I4 I 15

___-______ - Number of cases with skin tags

Pooled estimates, 89 degrees of freedom.

OPERATION AND NUMBER OF CASES

Submucosal Excision with Excision with Excision with Excision Low Ligation Primary Suture Clamp and Cautery

(28) (35) (16) (16) -

16 (57 per cent) 12 (34 per cent) 5 (31 per cent) 2 (12.5 per cent) ( 5 severe) (3 severe) (2 severe) (0 severe)

Table II.--PROPORTION OF PATIENTS WITH UNHEALED WOUNDS 6 WEEKS AFTER HAEMORRHOIDECTOMY

I OPERATION AND NUMBER OF CASES

Table III.--INCIDENCE OF FIBROSIS AND STENOSIS 6 MONTHS AFTER HAEMORRHOIDECTOMY ~


Suhmucosal Excision with Excision with Excision with PRESENCE ANU DEGREE OF FIBROSIS AND STENOSIS 1 Ex;;:? 1 Low g r i o n 1 Primaf;6kinm Clamp and Cautery 1 (16) _ I I

ImDassahle to fineer I 0 I 0 I 0 I 0 I 0 Passahle to finger-impassable to proctoscope I o I I* I I I I I I *

Passable to finger and proctoscope. but fibrosis

All cases -

palpable in wall ~ _ _ _ _ _ _ _ ~

* Associated symptoms present

Table IV.-INCIDENCE OF ANAL SKIN TAGS 6 MONTHS AFTER HAEMORRHOIDECTOMY

Table V.-RECURRENCE OF HAEMORRHOIDS WITH SYMPTOMS 6 MONTHS AFTER HAEMORRHOIDECTOMY

Suhmucosal Excision I (28)

Number of cases with recurrence 4 (14 per cent)

Saturc of symptom {Bleeding -'[-F Prolapse I


Excision with Excision with Excision with Low Ligation Primary Suture Clamp and Cautery

(35) I (16) I (16)

I ( ? percent) I o I 0

Excision with High Ligation

( 5 )

0 - -

WATTS E T AL.: HAEMORRHOIDECTOMY

The distribution of the I Z cases with palpable anal fibrosis or stenosis, relative to the type of haemorrhoidectomy performed, is given in Table IZZ.

Skin Tags.-Skin tags were present in 35 of the roo patients. They were particularly common after submucosal haemorrhoidectomy (see Table IV) , perhaps because of the smaller amount of external skin excised by this technique.

Two patients, after submucosal haemorrhoidectomy, developed a fibrous, mucosal-covered polyp, arising in each case at the level of the anorectal ring. One of these polyps, z cm. in length, repeatedly prolapsed at defaecation and has been excised at a further operation. The other, I cm. in length, has caused no symptoms and has not required treatment.

Recurrence of Piles.-On proctoscopic examination following any type of haemorrhoidectomy it is not uncommon to find that, when the patient bears down vigorously, the mucosa, between the intra-anal scars and above the sites of excision, bulges into the lumen. It is very difficult to evaluate this state of affairs in terms of haemorrhoidal recurrence, unless it is sufficient to result in mucosal protrusion at the anal orifice on withdrawing the proctoscope, or is associated with recurrent symptoms. In this inquiry, therefore, we have disregarded objectively demonstrated, symptomless piles and have accepted as significant recurrences only those patients with sizeable haemorrhoids accompanied by symptoms for which no other cause could be found. On this basis 5 patients were found to have recurrence; the original operations performed in these cases, and the nature of the recurrent symptoms, are shown in Table V. Two of the 4 cases with recurrence after submucosal haemorrhoidectomy have required in- jection treatment, and I has needed a further haemorrhoidectomy. For the recurrent case after excision with low ligation no further treatment has been necessary as the bleeding has been slight and infrequent.

DISCUSSION WOUND HEALING

These observations have demonstrated precisely the nature of the wounds produced in the anal canal by different forms of haemorrhoidectomy, and the way in which they heal. They show that some of the previous conceptions, based though they were on inadequate sources of information, are essentially correct, whilst others are seriously in error.

Ear ly State of the Anal Canal.- After Haenzorrhoidectomy with High Ligation.-Our

findings, admittedly in a small group of cases, have confirmed that this operation does leave very considerable raw areas in the anal canal, extending longitudinally up to the anorectal ring. In addition, these extensive wounds sometimes coalesce in the upper anal canal, resulting in the formation of a complete or nearly complete circumferential granulating area, as described by Milligan (1929).

After Haemorrhoidectomy with Low Ligation.-On the other hand, we have been quite unable to sub- stantiate the claim of Milligan and others (1937) that the low-ligation technique maintains a mucosal covering in most of the anal canal during the healing process. On the contrary, we have found that,

without exception, patients treated by this method of haemorrhoidectomy show, 10 days after operation, raw areas of varying width extending up to the top of the anal canal. We presume that the upper parts of these open wounds, above the pectinate line, are due to upward and lateral retraction of the mucosa when the ligatures are cut through. Indeed, immediately on completion of the operation, the ligatured pedicle can be observed to retract I-2cm. into the anal canal.

Despite the fact that some mucosa was always preserved between the three main piles at operation it was disappointing to find that, in one-third of the cases at 10 days after operation, one or more of these mucosal bridges were broken down, usually high in the anal canal, so that adjacent wounds had become confluent. No patient had developed a completely annular raw area, but the breadth of surviving mucosa remaining at this level often amounted to no more than one or two strips h-4 in. wide.

I t will be evident from these findings that the theoretical basis for the technique of haemorrhoidectomy with low ligature is quite invalid.

Haemorrhoidectomy with Immediate Suture.-This operation also signally failed to attain its theoretical objective in our hands. Almost invariably, by the tenth day, we found that the intra-anal wounds had separated, especially in their upper parts. Primary healing of the skin wounds was more common, and nearly 5 0 per cent of these wounds united satisfac- torily, though never more than one or two in any particular patient.

Submucosal Haemorrhoidectomy.-The submucosal technique was designed primarily to preserve the mucosa, so that it would cover the raw areas left after excision and high ligation of the vascular elements of the pile. Our misgiving was that these very thin and friable mucosal flaps might not survive in toro, particularly if damaged by diathermy coagulation of bleeding points during the operation. However, Parks (1956) has stated that on proctoscopic examination of some of his patients z weeks afterwards the wounds were all completely healed! We have to report that in our cases, 10 days after operation, most of the intra-anal wounds had closed, and the remainder were usually represented by narrow longitudinal strips of granulation tissue seldom more than t i n . wide. The external skin wounds were usually open, but were usually a little smaller at this stage than those resulting from ordinary haemorrhoidectomy with low ligation.

Haemorrhoidectomy with Clamp and Cautery.-It was reasonable to suppose that the linear eschars resulting from this operation would eventually separate, and that the mucosal edges might then retract away from one another leaving open wounds. But we did not anticipate that the amount of mucosal damage and sloughing would be quite so extensive. In many of the cases one or two of the mucosal strips between the piles were entirely destroyed, and in some of them there was complete circumferential loss of anal canal lining.

La ter Healing.-If the striking feature of the early review was the great amount of destruction of the lining of the anal canal found after several of the techniques employed, the remarkable aspect of the findings on later examination was the rapidity with


which the mucosa regenerated. In all patients, even those with the most extensive necrosis of the mucosal lining, this upper part of their wounds had completely healed by the time they were reviewed 6 weeks after operation.

Any delay in healing always related to the skin- lined part of the canal below the pectinate line, and

PER CENT PATIENTS WITH BRIDGE BREAKDOWN

(10 days)

n

2onU 0

PER CENT PATIENTS WITH FIBROSIS

( 6 months)

PER CENT PATIENTS WITH RECURRENCE

0- (6 months)

FIG. 6.--?‘he relationship between mucosal destruction in the anal canal at 10 days and the subsequent rates of anal canal fibrosis and haemorrhoid recurrence at 6 months, after various forms of haernorrhoidectomy.

to the perianal skin. Actually, there was little difference in the incidence of delayed skin healing after the four main operations on trial, except that external healing after the clamp and cautery method sometimes took a very long time.

Ultimate Fibrosis, Stenosis, and Recurrence. -We are hesitant to draw definite conclusions about the general incidence of these sequelae from such a small group of patients, followed up for only 6 months. None the less, some observations seem pertinent as the results so far show wide variations according to the type of operation employed. We have illustrated in Fig. 6 the frequency of palpable anal canal fibrosis, and of recurrence of piles, 6 months after the various operations. These findings are presented relative to the amount of mucosal destruction seen at the lo-day review, the latter being measured by the number of patients in each group who have lost one or more mucocutaneous bridges between the anal wounds.

Not unnaturally, there seems to be a correlation between the extent of mucosal damage and the incidence of subsequent fibrosis. But the surprising thing is that true stenosis was not much more frequent than we have found it to be in patients who had really extensive necrosis of anal canal lining, often with circumferential granulating wounds. Actually all the cases with complete mucosal destruction healed without producing significant stenosis at this level, although a few developed fibrous induration in the wall of the anal canal. These findings serve to illustrate the remarkable propensity for healing in the mucosal-lined part of the anal canal, where large denuded areas are presumably covered with epithe- lium before enough submucosal fibrosis is produced to lead to more than the slightest narrowing of thc anal canal. Indeed, in the 4 cases with genuine strictures (as measured by inability to pass a proctoscope) observed in these IOO patients, the site of maximal stenosis in each instance was at, or just below, the pectinate line.

I t also appears from these short-term results that the chance of recurrence of haemorrhoids depends to some extent on the amount of mucosa preserved. Already we have a 14 per cent true recurrence rate after 28 submucosal operations, the major attraction of which is the preservation of large mucosal flaps to cover the anal canal. Only I of the other 72 patients had a recurrence, according to our definition, and all these patients underwent an operation which we have shown to be associated with considerable superficial destruction in the anal canal. It might be argued that such a high recurrence rate, after the ‘mucosa sparing’ submucosal haemorrhoidectomy, reflects technical inadequacies in the performance of this rather difficult operation. We can only say, in reply, that all these operations were carried out in a unit in which much interest is lavished on rectal cases, and that the results in the hands of the average surgeon might be even less satisfactory.

Only a long-term follow-up study will confirm or refute our impression that ‘sparing’ of the mucosa in haemorrhoidectomy predisposes to recurrence, but reported long-term recurrence rates after the more destructive low-ligature operation (Soderlund, 1962; Bennett, Friedman, and Goligher, 1963 j and the clamp and cautery method (Smith, 1875; Cormie and McNair, 1959) have not even approached the incidence of early recurrence after the submucosal operation in this series.

POSTOPERATIVE PAIN Among laymen the surgical treatment of pilcs

carries a notorious reputation for the severity of postoperative pain, and not a few patients have long deferred a much-needed haemorrhoidectomy because of their apprehension on this score. Recognition of this fact has been a powerful stimulus to surgeons in introducing various modifications of technique for this operation, and few such technical innovations are unassociated with a statement that they help t o reduce the patient’s discomfort. Yet the evidence for these claims is often flimsy in the extreme, and is usually based on rather vague and often retrospective clinical impressions.

Our assessment of pain (see p. 809) has been carried out as a forward study; a prescribed set of criteria


has been applied in an identical manner to each case; and the evaluation has been performed, not by an individual, but by a panel of observers. We believe that the results obtained are fairly reliable, and certainly more so than in any other study of this subject previously reported.

The first point that emerges from Table Z is the apparent beneficial effect on postoperative pain of stretching the anal sphincters. This is shown in the group of 36 patients undergoing haemorrhoidectomy with low ligation. Nineteen patients of this group had a preliminary sphincter-stretch, and 17 had not. The results indicate that stretching of the sphincters, as a preliminary to this particular operation, secured a reduction in the severity of postoperative pain, which is not quite significant at the 5 per cent level.

Turning again to Table Z it will be seen that four groups of patients had different forms of operation combined with sphincter stretching. Comparison of these four groups shows that there was no obvious difference in the amount of pain experienced after excision with low ligation, after submucosal excision, and after excision with clamp and cautery, but that the method of excision with primary suture appeared to cause appreciably more pain than did these three operations.

In attempting to lessen the pain of haemorrhoidectomy, the method of primary closure would therefore seem to have nothing to offer in comparison with the other three operations. As between the latter, there would appear to be little to choose so far as the amount of postoperative discomfort is concerned, which is remarkable considering the great difference in the state of the anal wounds after these three operations. However, to meet possible criticisms of the way in which this comparative study has been conducted, we have considered it desirable to put this conclusion to the test of a more extended, properly controlled trial, which will soon be com- pleted and will be the subject of another paper.

SUMMARY A study of healing after five different forms of

haemorrhoidectomy (excision with high ligation, excision with low ligation, excision with primary suture, submucosal excision, excision with clamp and cautery) has been conducted on 104 patients.

Ten days after all forms of haemorrhoidectomy, except submucosal excision, extensive wounds were found in the anal canal up to the level of the anorectal ring. Breakdown of mucosal bridges between the intra-anal wounds was common, resulting in coales- cence of adjacent granulating wounds and sometimes

complete, or nearly complete, circumferential wounds. The most extensive destruction was observed after the clamp and cautery operation.

Despite these extensive intra-anal wounds, healing of the mucosal part of the anal canal proceeded rapidly, and was complete in all cases at 5-6 weeks. At this stage one or more skin wounds were unhealed in 15 patients, the proportion of patients with delayed skin healing being similar after all operations.

Six months after operation, 4 patients had developed fibrous strictures, at or below the pectinate line, sufficient to prevent passage of a proctoscope but not the finger; in addition 8 patients had palpable fibrous induration in the anal wall without significant narrowing. Fibrosis was commoner after the more destructive operations, but many patients with confluent intra-anal wounds healed without this develop- ment.

Recurrence of haemorrhoids with symptoms occurred in 4 of 28 patients within 6 months of submucosal excision, but in only I of 72 patients after other forms of haemorrhoidectomy. I t is suggested that preservation of the mucosa at operation may predispose to later recurrence.

Preliminary stretching of the anal sphincters at operation effected a significant reduction in pain after the low ligation and excision haemorrhoidectomy. There was no difference in pain following the submucosal, low ligation, and clamp and cautery operations. Excision with primary suture, however, was more painful than these three procedures.

REFERENCES ALLINGHAM, W., and ALLINGHAM, H. W. (1896), The

Diagnosis and Treatment of Diseases of the Rectum, 6th ed. London : Bailliere, Tindall & Cox.

ANDERSON, H. G. (1909), Brit. med. J., 2, 1276. BENNETT, R. c . , FRIEDMAN, M. w. H., and GOLIGHER,

CORMIE, J., and MCNAIR, T. J. (1959), Scot. med. y., CUSACK, J. W. (1846), Dublin quart. J. med. Sci., 2, 562. FARQUHARSON, E. L. (1962), Textbook of Operative

FERGUSON, J. A., and HEATON, J. R. (1959), Dis. Colo~t

MILES, W. E. (1919), Surg. Gynec. Obstet., 29, 497. MILLIGAN, E. T. C. (I929), Proc. R. Soc. Med., 23,

- - MORGAN, C. NAUNTON, JONES, L. E., and OFFICER,

MITCHELL, A. B. (I903), Brit. med. J., I , 482. PARKS, A. G. (1956), Brit. J. Surg., 43, 337. SMITH, H. (1875), Lancet, 2, 89, 124. SODERLUND, S. (1962), Acta chir. scand., 124, 444.

J . C. (1963), Zbid., 2, 216.

4, 571.

Surgery, 2nd ed., p. 725. London: Livingstone.

Rect., 2 , 176.

706.

R. (1937), Lancet, 2, 1119.

Healing and pain after haemorrhoidectomy

Documents