Classification of internal haemorrhoids: a discussion paper.

In the absence of an internationally agreed classification of haemorrhoids, evaluation of the ever increasing number of therapeutic options available is impossible. A new classification is proposed which relates symptoms to our knowledge of haemorrhoidal development, the use of which should provide a more accurate way of comparing treatments and in the evaluation of new techniques. The evidence for such a classification is presented in detail.

The results at nine to twelve years of the use of a hydroxyapatite-coated femoral stem.

We describe the clinical and radiological outcome of 100 consecutive total hip replacements in 86 patients using the JRI Furlong hydroxyapatite-coated femoral component. The follow-up was 100% at a mean of ten years (9.0 to 12.0). All 54 living patients (62 hips) were examined annually. Their mean Merle d'Aubigné score was 5.8/5.6/4.6. A total of 32 patients had died. Their notes at the last annual assessment recorded a similar outcome to that of the living patients. Radiographs showed no radiolucent lines around any femoral component and there were no signs of impending failure. One patient required excision arthroplasty, but the femoral component had been bonded satisfactorily. There have been no other revisions of femoral components. Our findings suggest that the hydroxyapatite coating gives a satisfactory prosthesis-bone interface which is preferable to any other system.

The aetiology and surgery of carcinoma of the anus, rectum and sigmoid colon in Crohn's disease. Negative correlation with human papillomavirus type 16 (HPV 16).

Anal and recto-sigmoid Crohn's disease may rarely be complicated by the development of local malignancy. Eight patients with this association were identified from the records of St. Mark's Hospital between 1947 and 1988 and two from The London Hospital. The aetiology of the malignancies and their surgical management have been examined. Eight patients had adenocarcinoma of the anus, rectum or sigmoid colon and two had squamous cell carcinoma of the anus. Synchronous dysplasia, adenomas and carcinomas were found in four of the eight patients with adenocarcinoma and a progression to malignancy which is analogous to that in ulcerative colitis is proposed for these cases via a 'dysplasia-carcinoma' or 'adenoma-carcinoma' sequence. In the four other patients with adenocarcinoma, the tumour arose within an area of Crohn's disease or in association with a chronic Crohn's fistula. In these four cases no dysplasia was found in the specimens at that time and long-standing infection is the only aetiological factor identified. Evidence for infection with human papillomavirus (HPV 16) was sought by DNA hybridisation of archival material but none was found in material from the Crohn's disease, fistulae or adeno- or squamous cell carcinomas. Anal and rectosigmoid Crohn's disease disguised the presence of malignancy, and diagnosis was usually delayed. All patients underwent excisional surgery and eight of the ten had all of the large bowel removed either as a single procedure or in stages. Only three of the eight cases of adenocarcinoma had early malignancies (Dukes' stage A) and the remaining five cases had locally advanced disease (Dukes' stage B).

Effect of abdominal Ivalon rectopexy on bowel habit and rectal wall.

Bowel habit in 57 rectal prolapse patients was assessed before and after abdominal Ivalon rectopexy. There was a significant (chi-square = 8.7, P less than 0.01) increase in prevalence of constipation from 30 percent before to 51 percent after surgery. There were two explanations for this increased constipation. It was mainly the result of a 28 percent increase in prevalence of constipation among patients who were incontinent before rectopexy. Incontinent prolapse patients were more likely to acquire a predictable bowel habit after rectopexy if they became constipated. There was also a small (7 percent) increase in prevalence of constipation among continent patients, which could be attributed to the rectopexy procedure. In a subgroup of 15 patients, rectal wall thickness after rectopexy was assessed by pelvic computed tomographic scan carried out before and after surgery, or at more than one year after surgery. There was a significant (t = 4.5, P less than 0.001) increase in rectal wall thickness by 24 weeks after rectopexy, compared with before operation. This increase was also seen in a further five patients undergoing abdominal rectopexy without Ivalon sponge, suggesting that it was a consequence of rectal mobilization rather than the Ivalon sponge. This increased rectal wall thickness may impede the passage of formed stool into the lower rectum and contribute to the increased constipation found after rectopexy.

A prospective study of infrared coagulation, injection and rubber band ligation in the treatment of haemorrhoids.

One hundred patients with non-prolapsing and one hundred with prolapsing haemorrhoids were allocated to receive conventional treatment (CT) by injection sclerotherapy or rubber band ligation, or infrared photocoagulation (IRC). Significantly more patients with nonprolapsing haemorrhoids were symptom free after IRC (81%) than CT (59%) at three months. (Chi2 = 4.4, p = 0.05). There was no significant difference in the outcome at 1 or 4 years. Likewise for prolapsing haemorrhoids, there was no significant difference in the outcome of IRC or CT at 3 months, one or 4 years. However, recurrence of prolapse was more common after IRC (54%) than rubber band ligation (RBL) (27%) at 1 year (Chi2 = 3.46, p less than 0.1). IRC was significantly less painful than CT (p less than 0.001). IRC is a safe, rapid, non-invasive alternative to CT, which is acceptable to the patient and give similar results, though RBL provides more rapid and longer lasting relief from prolapse.

Pelvic drainage after anterior resection of the rectum.

Thirty patients undergoing anterior resection of the rectum were randomized for postoperative drainage with either corrugated (N = 14) or corrugated plus Shirley sump drain (N = 16). Drainage volume per 24 hours was measured, and volume of residual pelvic fluid collection estimated by CT scan on the seventh postoperative day. A mean of 591.9 +/- 415.2 ml of fluid drained during the first seven postoperative days. There was a significant (P less than 0.01) increase in 24 hour drainage volume between the fourth and sixth postoperative days. There was no significant difference in drainage volumes between the two drainage methods. Residual pelvic fluid collection (median volume, 16 ml) was detected in 80 percent of patients at one week after operation. While this was larger (24 ml median) for the corrugated only group compared with the corrugated plus sump drain group (11.5 ml median), the difference was not significant. Fluid loss during the first postoperative week (616.6 +/- 424.2 ml) was significantly (P less than 0.05) less when the anastomosis was situated higher than 12 cm (294 +/- 192 ml) compared with 6 to 12 cm from the anal margin (496 +/- 210 ml), or after coloanal anastomosis (1077 +/- 432 ml). Residual pelvic fluid collection was larger after coloanal anastomosis (median, 33 ml; range, 1.5 to 90 ml) compared with 6 to 12 cm (median, 11 ml; range, 0-124 ml) or higher than 12 cm (median 9 ml; range, 0 to 16 ml) from the anal margin, but the difference was not significant. Drainage after anterior resection is important because large volumes of serosanguineous fluid collect, especially after resection of low tumors. Neither of the drainage methods tested in this study prevented persisting pelvic fluid collection at the seventh postoperative day.

Regional injection of 5-fluorouracil and starch microspheres in colorectal cancer.

Experiments were undertaken to examine the effect of degradable starch microspheres (DSM) on the local distribution of 5-fluorouracil (5-FU) given by mesenteric arterial injection. Tritiated 5-FU (200 microCi) was injected into the inferior mesenteric artery of 8 perfused specimens of human large bowel containing adenocarcinoma and the concentration of 5-FU in local tissues and venous effluent was measured by scintillation counting. Metabolites of 5-FU were detected by radiochromatography. DSM (6 X 10(7)/mean diameter 40 micron) were injected in suspension with 5-FU (n = 7) and separately: before (n = 7) and after 5-FU (n = 7). Specimens given DSM retained 20% more cytotoxic than specimens given 5-FU alone (p less than 0.05). The concentration of 5-FU in tissues next to the primary tumour was increased when DSM were given in suspension with 5-FU, but higher concentrations were obtained when DSM were given last (p less than 0.05). Low concentrations were produced when DSM were given first. Concentrations in tumour tissue were not significantly increased by DSM.

Rigid sigmoidoscopy: an evaluation of three parameters regarding diagnostic accuracy.

Rigid sigmoidoscopy is the method employed for examining the rectum and sigmoid colon in most general surgical clinics. Commonly, this is performed without any prior preparation of the bowel and with the patient in the left lateral (Sims') position. This study was designed to assess three factors relating to this method: preparation of the bowel; position of the patient; and the experience of the operator. The study shows that diagnostic accuracy without preparation is adequate in only 50 per cent of cases but, by employing a preliminary disposable (Fletcher's) enema, this can be improved to 80 per cent. Employing the knee-elbow position gave only slight advantage over the usual Sims' position and operator experience did not improve diagnostic accuracy once familiarity with the use of the sigmoidoscope had been achieved. The study underlines the importance of preparation as an essential prelude to sigmoidoscopy.

The immediate response to injection therapy for first-degree haemorrhoids.

Over an 8-month period, 100 consecutive patients undergoing sclerotherapy for first-degree haemorrhoids were issued with a questionnaire to assess responses to this treatment. Success was defined as complete cessation of bleeding at defaecation. The effect on bleeding was assessed at the end of 24 hours (99 responders) and 4 weeks later (98 responders): of 61 patients (62%) with no bleeding at 24 hours, only 40 (41%) remained symptom-free at 28 days post-injection. Twelve patients were treatment failures (either unchanged or increased bleeding post-injection). More than half the patients (n = 59) experienced pain related to the injection, which was severe in 9 cases. Although only 3 patients expressed complete dissatisfaction with the treatment they received, and overall 88% were either cured of bleeding or improved, the results suggest that critical judgment should be exercised before recommending the treatment to patients with minimal occasional bleeding due to first-degree haemorrhoids.

Complete rectal prolapse: the anatomical and functional results of treatment by an extended abdominal rectopexy.

Between 1970 and 1985 (inclusive), 66 patients presented with complete rectal prolapse; 59 (89 per cent) were treated by extended abdominal rectopexy. Forty-four patients (75 per cent) have been followed for more than 2 years: all cases were cured of their complete prolapse, no patients died, and major complications were few. Constipation (47 per cent) and incontinence (19 per cent) caused serious problems for many patients postoperatively.

The management of late radiation-induced rectal injury after treatment of carcinoma of the uterus.

Sixty-one of 1,418 (4.3 per cent) patients treated with radiation for carcinoma of the uterus from 1963 to 1983 had significant radiation-induced complications of the intestine develop which required a surgical opinion considering further management. Ninety-three per cent of these complications involved the rectum. Florid proctitis resolved within two years of onset in 33 per cent of the patients who were managed conservatively while 22 per cent of the patients died of disseminated disease within the same time period. Surgical treatment was eventually necessary in 39 per cent of the patients who were initially treated conservatively for radiation induced proctitis. Rectal excision with coloanal sleeve anastomosis produced a satisfactory result in eight of 11 patients with severe radiation injury involving the rectum. The incidence of radiation-induced and malignant rectovaginal fistula were similar (1 per cent), but disease-induced symptoms tended to occur earlier after primary treatment (a median of eight months) compared with radiation-induced symptoms (a median of 16 months).

Carcinoid tumours of the rectum and anus.

Carcinoid tumours of the anorectum comprise only 0.1 per cent of all rectal tumours. Their behaviour and treatment are still debatable. We identified 35 cases from the records of St Mark's Hospital; 31 were benign and 4 were malignant. Benign tumours were of three distinct histopathological patterns. All benign tumours were cured by local treatment. Malignant tumours can be identified clinically; the only consistent microscopic feature was invasion of the muscularis propria. All patients with malignant tumours died, despite radical surgical treatment; lymph node metastases were invariably present in the operation specimens. Because radical treatment for malignant carcinoid tumours of the anorectum is largely unsuccessful, a flexible surgical approach is advocated.

'D' excision for sacrococcygeal pilonidal sinus disease.

A new procedure is described for treating pilonidal sinus by an excision and primary suture technique, and the results reported in 30 consecutive patients so treated--28 with chronic sinuses and 2 with an acute abscess. Seventeen patients (Group 1) had had no previous surgery, while 13 (Group 2) had had multiple previous operations. A total of 24 patients (80%) healed after the operation, their mean hospital stay being 16 days. In Group 1 the success rate was 88% with a mean hospital stay of 15 days; in Group 2 the comparable figures were 69% and 17 days. After additional procedures (usually curettage) all patients healed.

Has the incidence of radiation-induced bowel damage following treatment of uterine carcinoma changed in the last 20 years?

Radiation-induced bowel damage occurred in 4.3% of patients treated primarily by irradiation for uterine carcinoma during the period 1962-1982. There has been a progressive rise in the incidence of radiation damage and radiation-induced rectovaginal fistula during this 20-year period. Radiation from intracavitary sources was a contributory factor in 92% of injured cases. The rising incidence of bowel damage in our patients may be due to an increase in the number of patients receiving a high rectal dose from the intracavitary source. There was a significantly (P less than 0.01) higher incidence of radiation injury in cases of cervical carcinoma compared to endometrial carcinoma. This was because cervical carcinoma tended to present at a more advanced stage than endometrial carcinoma and was more frequently treated with combined external and intracavitary irradiation. There was no significant increase in the incidence of complications among patients undergoing hysterectomy.

The results of surgical treatment of cancer of the rectum by radical resection and extended abdomino-iliac lymphadenectomy.

At St Mark's Hospital survival after radical surgery for cancer of the rectum has not changed significantly over the past 30 years. The technique of extended abdomino-iliac lymphadenectomy was developed in an attempt to improve prognosis in patients considered to have particularly unfavourable tumours. Between 1960 and 1981 the technique was used in 75 patients with a single adenocarcinoma of the rectum. Two patients died postoperatively and 52 patients developed complications; a mortality and morbidity similar to those seen after conventional surgery at this hospital. Five-year survival rate showed no improvement over that achieved by conventional techniques; disappointingly this was also the case for patients with Dukes' C1 tumours. The results of this study suggest that an improvement in survival in patients with cancer of the rectum is unlikely to be achieved by any extension of conventional radical surgery.