INTRODUCTION. British Journal of Obstetrics and Gynaecology July 2000, V01107, pp

Transcription

1 British Journal of Obstetrics and Gynaecology July 2000, V01107, pp The impact of adhesions on hospital readmissions over ten years after 8849 open gynaecological operations: an assessment from the Surgical and Clinical Adhesions Research Study Adrian M. Lower Consultant (Gynaecology), t Robert J. S. Hawthorn Consultant (Gynaecology), $Harold Ellis Emeritus Professor (Surgery), The late Fiona O Brien Senior Information Scientist, acot Buchan Directol; qalison M. Crowe Director For the complete list of the SCAR Study Steering Group please see page 861 *Fertiliq Centre, St. Bartholomew s Hospital, London; tdepartment of Gymecology, Southern General NHS Trust, Glasgow; $Department of Anatomy, Cell and Human Biology, The Guy s King s College and St. Thomas School of Biomedical Sciences, London; $Information and Statistics Division, NHS in Scotland, Edinburgh; vtrategen, Disease and Therapy Management, Basingstoke Objective To investigate the epidemiology of, and the clinical burden related to, adhesions following gynaecological surgery. Population The Scottish National Health Service Medical Record Linkage Database was used to define a cohort of 8849 women undergoing open gynaecological surgery in Methods All readmissions for potential adhesion related disease in the subsequent 10 years were reviewed. Main outcome measures Readmissions and the degree of adhesion involvement gave an indication of clinical burden and workload. The rate of readmission following the initial surgery determined the relative risk of disease related to adhesions. Results Two hundred and forty-five (4.5%) of 5433 readmissions following open gynaecological surgery were directly related to adhesions. 34.5% of patients were readmitted, on average 1.9 times, for a problem potentially related to adhesions or for further intra-abdominal surgery that could be complicated by adhesions. Readmissions related to adhesions continued throughout the 10 year period of the study. The overall rate of readmission was 64-0/100 initial operations. For readmissions directly related to adhesions, the rate was 2.9/100 initial operations. Operations on the ovary had the highest rate directly related to adhesions (7-5/100 initial operations), with an overall rate of readmission of 106.4/100 initial operations. Conclusions Despite the conservative approach taken in this study, the clinical burden, workload and relative risk of readmissions related to adhesions following open gynaecological surgery was considerable. Post-operative adhesions have important consequences for patients, surgeons and the healthcare system. These results emphasise the need for more effective strategies to prevent adhesions. INTRODUCTION Post-operative adhesions have long been recognised as a complication of general and gynaecological surgery. Indeed in 1872 Bryant reported a case of fatal intestinal obstruction due to adhesions following removal of an ovarian tumour. As the frequency of abdominal and pelvic surgery has increased, the incidence of pelvic adhesions has risen in direct proportion*. The complications of adhesions largely have been seen and treated by general surgeons, but with the Correspondence: Mr A. Lower, Fertility Cenbe, St Bartholomew s Hospital, London EC17BE, UK. increase in the surgical treatment of gynaecological disease, especially of infertility, the problems have become more evident to gynaecologists. Post-operative adhesions occur in 60%90% of women who have undergone major gynaecological surgery3. The morbidity and economic costs are recognised as substantial. Adhesions are the largest single cause of intestinal obstruction, accounting for 30%-41% of all cases requiring further surgery4 and 65%-75% of those cases with small bowel obstruction4v5. Estimates suggest that 15%-20% of cases of infertility in women are secondary to adhesions6. Furthermore, despite long-standing controversy regarding 0 RCOG 2000 British Journal of Obstetrics and Gynaecology 855

2 856 A. M. LOWER ET AL. the association between adhesions and pelvic pain, there is increasing evidence to support such a relationship. DiZerega7 reports a review of 11 studies of almost 1000 patients showing that adhesions were the most common pathology in patients with chronic pelvic pain (40%), although 25% had no pathology. Evidence is also accumulating that surgical division of adhesions leads to a reduction of pain. Analysis of three studies including 11 3 women examining the effect of adhesiolysis on chronic pelvic pain indicated significant benefits in 80% of cases. Post-operative adhesions are clearly recognised as a consequence of surgery, and there has been research into the morbidity and consequent costs, but most of the work previously published focuses only on surgical intervention for intestinal obstruction and comes mainly from small studies. The epidemiology of adhesions as an iatrogenic disease has not been fully described, nor has the burden of nonoperative management which forms an essential part of treatment. The Surgical and Clinical Adhesions Research (SCAR) Study was set up to investigate the epidemiology and burden of disease caused by post-operative adhesions by creating the world s largest verified database to identify the number of readmissions to hospital following abdominal and pelvic surgery. This analysis determines the consequences of post-operative adhesions through well documented clinical outcomes for in-patient admissions following open gynaecological surgery. METHODS The study used data from the Scottish NHS Medical Record Linkage Database. This is one of the largest patient linked datasets in existence. It contains details of all inpatient and day-case hospital admissions within Scotland since 1981 (excluding maternity and psychiatric admissions). A major advantage in the analysis of this database is the demographics of Scotland, a geographically self-contained country which has a national health service with centralised reporting. The population is relatively stable at 5.1 million and annual migration is less than 1%8. Thus, it provides an ideal model on which to undertake epidemiological research. The detailed methodology for the SCAR study is outlined elsewhere9. The initial surgery was categorised according to anatomical region: fore gut and related abdominal organs as upper abdomen ; mid and hind gut (encompassing intestinal surgery below the stomach) as lower abdomen ; and the female reproductive tract. This study focuses on women undergoing open surgery on the female reproductive tract in 1986 and their follow up over a 10-year period. All subsequent hospital readmissions for disease related to adhesions, or for further abdominal or pelvic surgery which could be complicated by adhesions, were examined. Women with a history of abdominal or pelvic surgery in the five years prior to 1986 have a higher potential for readmission related to adhesions9 and were excluded from subsequent analysis (Fig. 1). Only the readmissions of the remaining women undergoing open surgery on the female reproductive tract were further analysed, ensuring that a conservative approach was adopted in estimating the clinical burden and workload secondary to post-operative adhesions. The study was directed by a multidisciplinary steering group of gynaecological and general surgeons, and a health economist. The initial surgical procedures likely to result in the development of adhesions were defined using the Office of Population, Censuses and Surveys surgical codes (OPCS-3). The subsequent readmissions (operative and nonoperative) following the initial surgery were identified using the OPCS-3 and OPCS-4 surgical codes and the (ICD-9) diagnostic codes. All initial surgery unlikely to result in adhesions and surgery outside the abdominal and pelvic cavity were excluded. Similarly, all readmissions which were clearly not related to adhesions in the abdominal or pelvic cavity or unlikely to be complicated by abdominal or pelvic adhesions were excluded. As shown in Table 1, the readmissions were allocated to one of three categories: 1. Directly related to adhesions; 2. Possibly related to adhesions; 3. Open or laparoscopic abdominal or pelvic reoperations that could be complicated by adhesions. These categories were mutually exclusive. Thus, each readmission was allocated to only one category by defining the extent of adhesion involvement using specific surgical codes (OPCS-3 or OPCS-4) in combination with defined diagnostic codes (ICD-9) for confirmation. Nonoperative readmissions required a diagnostic code (ICD-9) for confirnation. This process ensured no double counting and was deliberately conservative. Readmissions categorised as directly related to adhesions were most closely analysed. Other readmissions considered as possibly related to or complicated by adhesions had a large impact and are discussed in the paper. Readmissions were broadly assessed in two ways. The actual numbers and percentage of readmissions categorised as directly or possibly related to adhesions or reoperations that could be complicated by adhesions gave information on the actual clinical burden and workload associated with adhesions. Assessment of the rate of readmission per 100 initial surgical procedures gave information on the relative risk of readmissions related to adhesions following surgery, which is useful in planning strategies to prevent adhesions. The study also investigated the surgical specialty to which the

3 ADHESION-RELATED READMISSIONS AFTER OPEN GYNAECOLOGICAL SURGERY 857 Total SCAR population n = 54,380 ~ ~~ Undergoing open surgery on female reproductive tract n = 10,922 Female reproductive tract open surgery cohort n = outcomes Excluded patients undergoing surgery in previous 5 years n = 2433 (52.2%) (47.8%) n = 2931 n = 5558 (34.5%) Fig. 1. Women undergoing surgery on the female reproductive tract. readmissions presented for treatment of problems related to adhesions. The accuracy of the data used in the study was ensured through considerable training and monitoring of the professional coders. The data held by the National Health Service in Scotland are validated at a hospital level by an annual 1% audit of local hospital clinical records compared with data held centrally'o~". These processes resulted in a highly accurate data set which was then independently and rigorously scrutinised by the steering group for the purpose of the SCAR study. The methods of the study and the processes of coding and capturing information on the linked dataset ensured that the data would provide a conservative interpretation of the problem9. RESULTS Open surgery on the reproductive tract was performed in 10,922 women in Of these, 2433 had a history of abdominal or pelvic surgery in the previous five years and were excluded. The open surgery cohort was comprised of 8489 women (Fig. 1) in whom further analysis was undertaken. Of these women, (34.5%) had a total of 5433 readmissions over the next 10 years, a mean of 1.9 readmissions per woman. There was variation in the number of readmissions per woman when the site of surgery was considered (Table 2). A higher percentage of women was readmitted following surgery on the ovary or fallopian tubes (48.1% and 41.2%, respectively). The majority of women were readmitted on one occasion only, although a significant number had between two and five readmissions for a complication that was either directly related to adhesions, possibly related, or for further surgery that would be complicated by adhesions (Table 3). Table 1. Hospital outcome readmission categories. Outcome category Directly related to adhesions Possibly related to adhesions 'Reoperations' Hospital outcome readmissions Adbesiolysis operations Nonoperative readmissions for adhesions Adhesiolysis operations on the female reproductive tract Gynaecological operations possibly related to adhesions Abdominal operations possibly related to adhesions Nonoperative readmissions possibly related to adhesions Abdominal and pelvic procedures that could be complicated by adhesions

4 ~ ~~~ ~ 858 A. M. LOWER ET AL. Table 2. Readmissions by site of surgery. Site of initial No. of women undergoing No. of women No. of Women Average per open surgery initial open surgery readmitted readmissions readmitted (%) woman ovary 624 Fallopian tubes 1171 Uterus 6616 Vagina 78 TOTAL The categories of readmission are shown in Table 4, allowing assessment of the actual burden of disease related to adhesions. One hundred and sixty-four abdominal surgical interventions for disease directly related to adhesions were identified over the 10-year study, of which 26 cases had small bowel obstruction. A further 57 nonoperative readmissions directly associated with adhesions were identified. In addition, there were 24 readmissions for gynaecological adhesiolysis. This represents 4.5% of all readmissions following open surgery on the female reproductive tract. Of the readmissions following open surgery on the female reproductive tract, 2479 (45.6%) were operative or nonoperative readmissions which were classified as possibly due to adhesions (Table 4). This group included 1201 readmissions for nonoperative Table 3. Frequency of readmissions per woman over 10 years. Values are given as n (%). No. of readmissions TOTAL No. of women 1675 (57.15) 1159 (39.54) 89 (3.04) 8 (0.27) 2931 (100) treatment of symptoms suggestive of small bowel obstruction or other disease related to adhesions, including pelvic pain which accounted for 22.1% of all readmissions. There were 748 readmissions for further gynaecological surgery and 530 readmissions for abdominal surgery considered possibly related to adhesions, including ovarian surgery, removal of uterine adnexae, diagnostic laparoscopy, hydrotubation or exploratory laparotomy. The burden of readmissions for treatment possibly related to adhesions vaned considerably, depending on the site of surgery. The greatest number of readmissions followed surgery on the fallopian tubes or ovary. A total of 2709 readmissions (49.9%) were for other abdominal or pelvic surgery potentially complicated by the presence of adhesions; 56.8% of all readmissions following initial open surgery on the uterus were in this category. The rate of readmission, defined as the number of readmissions during the 10-year follow up period per 100 initial procedures in 1986, provides an indicator of the relative risk of problems related to adhesions (Table 5). Overall a rate of 64.0 readmissions per 100 initial procedures was seen over the 10 years, with a rate of 2.9 per 100 initial procedures directly related to adhesions. The rate of readmission depended on the site of the initial surgery. Operations on the ovary had the Table 4. Readmissions following open surgery on the female reproductive tract. Values are given as n (%), unless otherwise stated. DRA = directly related to adhesions; PRA = possibly related to adhesions; PCA = potentially complicated by adhesions. Readmissions (DRA) n = 245 Readmissions (PRA) n = 2479 Repeat Site of initial Gynaecological Gynaecological Abdominal Nonoperative surgery (PCA) Total open surgery (n) Adhesiolysis* Adhesions** adhesiolysis surgery surgery readmission n = 2709 n = 5433 Ovary (624) 29 (4.4) 10 (1.5) 8 (1.2) 141 (21.2) 56 (8.4) 187 (28.2) 233 (35.1) 664 Fallopiantubes (1171) 28 (3.0) lo(1.1) 8 (0.8) 240 (25.4) 89 (9.4) 260 (27.6) 308 (32.7) 943 Uterus (6616) 103 (2.8) 37 (1.0) 8 (0.2) 362 (9.6) 377 (10.0) 738 (19.6) 2134 (56.8) 3759 Vagina (78) 4 (6.0) 0 (0.0) 0 (0.0) 5 (7.5) 8 (11.9) 16 (23.9) 34 (50.7) 67 TOTAL (8489) 164 (3.0) 57 (1.0) 24 (0.4) 748 (13.8) 530 (9.8) 1201 (22.1) 2709 (49.9) 5433 (100) *With or without small bowel obstruction. **With or without small bowel obstruction treated nonoperatively.

5 ADHESION-RELATED READMISSIONS AFTER OPEN GYNAECOLOGICAL SURGERY 859 Table 5. Rate of readmissions per 100 initial procedures over 10 years. Numbers in brackets are numbers of readmissions. DRA = directly related to adhesions; PRA = possibly related to adhesions; PCA = potentidy complicated by adhesions. Rate of readmissions (DRA) n = 245 Rate of readmissions (PRA) n = 2479 Repeat Site of initial Gynaecological Gynaecological Abdominal Nonoperative surgery (PCA) Total open surgery (n) Adhesiolysis* Adhesions** adhesiolysis surgery surgery treatment n = 2709 n = 5433 Ovary (624) 4.6 (29) 1.6 (10) 1.3 (8) 22.6 (141) 9.0 (56) 30.0 (187) 37.3 (233) (664) Fallopiantubes (1171) 2.4 (28) 0.8 (10) 0.7 (8) 20.5 (240) 7.6 (89) 22.2 (260) 26.3 (308) 80.5 (943) Uter~s (6616) 1.6 (103) 0.5 (37) 0.1 (8) 5.5 (362) 5.7 (377) 11.1 (738) 32.2 (2134) 56.8 (3759) Vagina (78) 5.1 (4) 0 (0) 0 (0) 6.4 (5) 10.2 (8) 20.5 (16) 43.6 (34) 86.0 (67) TOTAL(8489) 1.9 (164) 0.7 (57) 0.3 (24) 8.8 (748) 6.2 (530) 14.1 (1201) 32.0 (2709) 64.0 (5433) *With or without small bowel obstruction. **With or without small bowel obstruction treated nonoperatively. highest rate of readmissions directly related to adhesions (7.5 per 100 initial procedures) with an overall rate for all causes of One of the most important aspects of the study is the analysis of readmissions over time for problems related to adhesions (Fig. 2). Of all readmissions, 16% were seen in the initial post-operative year, but readmissions continued steadily throughout the period of the study for all categories, with no evidence of decline. An analysis of readmission by specialty showed that only 2379 of the 5433 readmissions (43.8%) were to gynaecology with 1696 (31.2%) readmitted to general surgery. Following open surgery on the female reproductive tract, 52.2% of readmissions directly related to adhesions were to gynaecology with 26.1?& to general surgery (Table 6). An analysis of the total SCAR dataset examined the rate of all readmissions for gynaecological adhesiolysis and showed slightly higher rates following open surgery on the appendix or abdominal wall than after surgery on g 80 v- j I No. of years after initial procedure Fig. 2. Readmissions over time. [----I directly related to adhesions n = 239 [+] possibly related to adhesions n = 2413; [-&I repeat surgery complicated by adhesions n = 2629; [-I total n = Table 6. Readmission to surgical specialties following open gynaecological surgery. Values are given as n (%). Directly Readmissions Reoperations related Possibly related complicated to adhesions to adhesions by adhesions Surgical specialty (n = 245) (n = 2479) (n = 2709) General surgery 64 (26.1) 738 (29.8) 894 (33.0) Gynaecology 128 (52.2) 1463 (59.0) 788 (29.1) Urology 35 (14.3) 58 (2.3) 278 (10.3) General medicine 8 (3.3) 101 (4.1) 397 (14.7) Other 10 (4.1) 119 (4.8) 352 (12.9) the female reproductive tract, although the numbers were small (Table 7). Surgery at other sites in the lower abdomen showed no readmissions for gynaecological adhesiol y sis9. DISCUSSION This is the first study in a population of the clinical burden of post-operative adhesions following open surgery on the female reproductive tract. Previous research has been limited to retrospective or prospective analyses of clinical outcomes following surgery on small groups of patients, or analyses of patients presenting in a particular setting (e.g. patients with small bowel obstruction). The SCAR study, however, examined all readmissions to hospital in Scotland related to adhesions over a 10- year period following the initial abdominal or pelvic surgery in The Scottish National Health Service s Medical Record Linkage Database proved an excellent resource for undertaking this research. In addition, the demographics of Scotland are such that with centralised health reporting and a stable population* the data provide an excellent model for use in estimating the burden of disease and the planning of resources, as regards the

6 ~ 860 A. M. LOWER ET AL. Table 7. Comparison of rates of gynaecological readmissions. Values are given as n or rate* (nmdm,snon,). Total number of female Rate of readmissions Rate of readmissions for patients undergoing for gynaecological gynaecological surgery initial surgery adhesioly sis possibly related to adhesions Appendix (n = 5768) Abdominal wall (n = 2761) Other sites lower abdomen (n = 4055) Female reproductive tract (n = 8489) (9) 0.43 (7) 0 (0) 0.27 (24) 9.7 (254) 9-0 (148) 1.7 (36) 8.8 (748) *Per 100 initial procedures over 10 years. population and also in hospital. The results of this study can be extrapolated to other similar populations. Only women undergoing open surgery on the female reproductive tract in 1986 were considered in this analysis. Over 22% of women undergoing such surgery were excluded because they had had abdominal or pelvic surgery in the previous five years. These women have been shown to have a significantly higher rate of readmission (Fig. 1). Our study was performed on a cohort of women who were unlikely to have had any previous abdominal or pelvic surgery and thus had little chance of having any adhesions. The results of our study are likely to under-estimate the problem of post-operative adhesions. The study showed that following open surgery on the female reproductive tract over one in three women were readmitted on average 1.9 times during the 10-year period of the study. We consider this a very conservative interpretation of the burden of adhesions for the following reasons: 1. An explicit reference to adhesions in either the OPCS surgical procedure or ICD diagnostic codes was required to categorise a readmission as directly related to adhesions; 2. The ICD-9 diagnostic coding system has few precise references for adhesions and so many readmissions for nonoperative treatment of adhesions may have been included in the category possibly related to adhesions, with the ICD-9 codes recording symptoms such as abdominal pain, vomiting, nausea or gastrointestinal strangulation; 22.1 % of readmissions were in this category (Table 4); 3. Only seven confirmatory OPCS-3 and OPCS-4 codes specific for the surgical treatment of gynaecological adhesions were used to identify direct adhesion-related gynaecological readmissions, The remaining gynaecological surgical readmissions were categorised into gynaecological surgery possibly related to adhesions, which accounted for 748 readmissions, 1343% of the total (Table 4). It is recognised that pelvic inflammatory disease and endometriosis are additional potential causes of adhesions. This study has excluded women who have abdominal or pelvic surgery (open or laparoscopic) within the previous five years. Any impact of pelvic inflammatory disease or endometriosis is likely to be small. There may have been a small number of women requiring surgery for acute pelvic inflammatory disease, but we do not consider this to have any significant impact on the overall burden of readmissions related to adhesions. Even with this cautious methodology and interpretation, almost one in 22 readmissions (for operative and nonoperative treatment) following open surgery on the female reproductive tract were directly related to adhesions. This rate varied with the actual site of surgery, to the extent that while there were one in 25 readmissions following surgery on the uterus, the rate increased to one in 14 following surgery on the ovary. This variation with site of initial surgery was seen in all readmission groups. Initial surgery higher up the female reproductive tract resulted in a greater burden of readmission possibly because these women wished to conserve their fertility and were more likely to be readmitted for investigation of this. Whereas the clinical burden of adhesions, as estimated from the numbers of readmissions, provides a useful perspective of the actual surgical and medical workload associated with adhesions, analysis of the rate of readmission provides an indicator of relative risk. This is particularly useful for surgeons and health planners when initiating adhesion prevention strategies, and this information is also important to women in making decisions about their surgery. Surgery on the ovary and fallopian tubes appears to carry the highest relative risk for readmission for problems directly related to adhesions, with an overall rate of readmission of and 80.5 per 100 initial procedures, respectively (Table 5). It is clear that there is an important broader risk that warrants specific interest and action. Fifteen percent of readmissions for problems directly related to adhesions were in the first year, a figure comparable to that seen previously in women readmitted for intestinal obstruction12. However, most importantly, readmissions continued steadily over the 10 years studied, showing no decline with time. 0 RCOG 2000 Br J Obsret Gynaecol 107,

7 ADHESION-RELATED READMISSIONS AFTER OPEN GYNAECOLOGICAL SURGERY 861 Surgery undertaken on patients today will have considerable long term repercussions. While small bowel obstruction attracts most attention in studies of postoperative adhesions, our study demonstrates that the magnitude of the overall workload related to adhesions in reoperative surgery should not be underestimated. Nearly 50% of all readmissions were for reoperations that could potentially be complicated by the presence of adhesions. In addition, the 1466 readmissions (27%) for abdominal and gynaecological surgery possibly related to adhesions and abdominal and gynaecological adhesiolysis could also be complicated by adhesions (Table 4). Studies have shown that up to 93% of patients who have undergone a laparotomy will develop adhesions5.13 and 55%-100% of women who have undergone gynaecological laparoscopic surgery will develop post-operative adhesions. Work by one of the SCAR steering group (B. J. Moran) has shown that in a series of otherwise routine laparotomies, the presence of adhesions adds an average of 24 minutes to the operating timei4. Other research has shown a 19% risk of bowel perforation related to adhesions during laparotomy15. It has been suggested that, due to the less invasive nature of laparoscopic surgery, post-operative adhesions occur less frequently than following open surgery. Adhesions themselves may make laparoscopic procedures more hazardous, increasing the risk of damage to major organs and bowel perforation. Thus the effect of adhesions from previous surgery on the workload and turnover of patients is considerable, let alone the effect on the successful outcome of the initial treatment. Twenty-two percent of readmissions were for nonoperative treatment of conditions possibly related to adhesions, predominantly abdominal and pelvic pain (Table 4). While the study cannot identify how many of these readmissions were directly the result of adhesions, previous studies have shown that 20%-50% of women with chronic pain have pelvic adhesions 6. Analysis of readmissions by specialty shows that surgery by one specialty often resulted in treatment by another specialty: 52.2% of readmissions directly related to adhesions were to gynaecology, with the remainder readmitted to other specialities, particularly general surgery and urology (Table 6). It is well recognised that surgery on the female reproductive tract causes intra-abdominal adhesions and small bowel obstruction as well as pelvic adhesions. This study has also confirmed previous suspicions that, in women, appendicectomy appears to carry a risk of gynaecological adhesions similar to that seen following gynaecological surgery. With the pressure to improve outcomes and the throughput of patients in hospitals, reduction of postoperative adhesions should be a priority. With the prob- lem continuing steadily over the 10 years studied, it is obvious that gynaecological surgery today is going to have considerable long term repercussions for surgeons, other clinicians and the health system as a whole, as well as for women. CONCLUSION This study emphasises the need for improved treatments to prevent adhesions, for there is no doubt that both morbidity and workload due to adhesions are considerable. Recent advance in the prevention of adhesions need to be further evaluated. Some substances have been shown in randomised trials to reduce the development of post-operative adhesion^^'.'^, and they should therefore be used more often. The results of the SCAR study demonstrate the magnitude of the problem and give useful information concerning women likely to benefit from these treatments, such as those undergoing surgery of the ovary and fallopian tubes, and women admitted with symptoms related to adhesions, including small bowel obstruction. Acknowledgements The authors would like to thank the Information and Statistics Division of the NHS in Scotland for allowing use of the Medical Record Linkage Database; Ms M. McCleod and Ms S. Houston, who provided input to the work and particularly validation of the dataset; and Mr A. Vos at Genzyme for his valuable input and critique of the work. The study received funding from Genzyme Therapeutics, Naarden, The Netherlands, Specifically we also wish to acknowledge the important contribution to this paper of one of our fellow authors, Fiona O Brien, who died tragically after a short illness prior to publication. The SCAR Study Steering Group Professor H. Ellis (The Guy s, King s College and St. Thomas School of Biomedical Sciences); Mr A. M. Lower (St. Bartholomew s Hospital); Dr R. J. S. Hawthorn (Southern General NHS Trust); Professor A. McGuire (City University, London and Oxford Health Economics Group); Mr D. Menzies (Colchester District General Hospital); Mr B. J. Moran (North Hampshire Hospital); Mr M. C. Parker (Joyce Green Hospital); Mr J. N. Thompson (Chelsea and Westminster Hospital); Mr M. S. Wilson (Macclesfield District General Hospital and Stepping Hill Hospital); the late Ms F. O Brien (Information and Statistics Division, NHS in Scotland); Ms A. M. Crowe (Strategen, Disease and Therapy Management); Mr s. Buchan (Strategen, Disease and Therapy Management).

8 862 A. M. LOWER ET AL. References 1 Bryant T. Clinical lectures on intestinal obstruction. Med Times Gaz 1872; 1: Ellis H. The cause and prevention of postoperative intraperitoneal adhesions. Surg Gynecol Obsfef 1971 ; Monk BJ, Berman ML, Montz FJ. Adhesions after extensive gynecologic surgery: clinical significance, etiology and prevention. Am J Obstet Gynecol1994; 170: Menzies D. Postoperative adhesions: their treatment and relevance in clinical practice. Ann R Coll Surg Engll993; 75: Ellis H. The magnitude of adhesion related problems. Ann Chir Gynaecoll998; 87: Soules MR, Dennis L, Bosarge A, Moore DE. The prevention of postoperative pelvic adhesions: an animal study comparing barrier methods with dextran 70. Am J Obstet Gynecol 1982; 143: di Zerega GS. Biochemical events in peritoneal tissue repair. Eur J Surg 1997; 163 (Suppl577): Social Trends, No. 25. London: Central Statistical Office, Ellis H, Moran BJ, Thompson JN et al. Adhesion-related readmissions after abdominal pelvic surgery: a retrospective cohort study. Lancet 1999; 353: Kendrick S, Clarke J. The Scottish record linkage system. Health Bull 1993; 51: Harley K, Jones C. Quality of Scottish morbidity record (SMR) data. Health Bull 1996; 54: Raf LE. Causes of abdominal adhesions in cases of intestinal obstruction. Acta Chir Scand 1969; 135: Menzies D, Ellis H. Intestinal obstruction from adhesions: how big is the problem? Ann R Coll Surg Engl1990; 72: Coleman MG, McLain AD, Moran BJ. The impact of previous surgery on the time taken for incision and division of adhesions during laparotomy. Dis Colon Rectum (in press). 15 van der Krabben, Dijkstra FR, Nieuwenhuijzen M et al. Morbidity and mortality of inadvertent enterotomy during adhesiotomy. Br J Surg 2000; 87: Monk BJ, Berman ML, Montz FJ. Adhesions after extensive gynecologic surgery: clinical significance, etiology and prevention. Am J Obsfet Gynecoll994; 170: Diamond MP on behalf of The Seprafilm Adhesion Study Group. Reduction of adhesions after uterine myomectomy by Sepraflm membrane (HAL-F): a blinded, prospective, randomized, multicenter clinical study. Fertil Sted 1996; 66: Becker JM, Dayton MT, Fazio VW et al. Prevention of postoperative abdominal adhesions by a sodium hyaluronate-based bioresorbable membrane: a prospective, randomized, double-blind multicenter study. JAm Coll Surg 1996; 183: Accepted 19 January 2000