|HOME||HELP||FEEDBACK||BROWSE ARTICLES||BROWSE BY AUTHOR|
clinmed/2000020012v1 (April 21, 2000)
Contact author(s) for copyright information
TOWARDS EVIDENCE-BASED CIRCUMCISION OF ENGLISH BOYS
AMK RICKWOOD, FRCS, Consultant Urological Surgeon
SE KENNY, BSc, FRCS, Lecturer in Paediatric Surgery
SC DONNELL, FRCS, Consultant Paediatric Surgeon
Department of Urology
Alder Hey Children’s Hospital
Liverpool, L12 2AP
Pathological phimosis, usually due to Balanitis Xerotica Obliterans, represents the one absolute indication for circumcision and recurrent balanoposthitis an occasional, relative indication.
Pathological phimosis affects less than 1% of boys and scarcely ever those under 5 years of age.
Although the rate of circumcision of English boys has declined by 42% during the 1990's, the current rate remains upwards of twice that to be expected of the incidence of preputial pathology.
The principal deficiency in current English practice is over-diagnosis of phimosis and especially so in boys less than 5 years of age.
The present NHS rate of medical circumcision would have 3.8% of English boys undergoing the procedure: an attainable, evidence-based, target figure would be no more than 2%.
Objective To assess the extent to which recent trends in medical circumcision of English boys match the incidence and age-distribution of preputial pathology.
Design Analysis of circumcision rates both overall and in various age range. Analysis of the age-distribution of preputial pathology among boys referred for circumcision.
Setting England (1984-98), former Mersey Region (1975-97), Liverpool Children’s Hospitals (1975-97), BUPA subscribers (1992-7).
Subjects Boys aged 0 -14 years.
Results Crude circumcision rates in the mid-1980's were geographically uniform (»4.2/000 boys/annum): the most recent corresponding rates are England 2.5, BUPA subscribers 2.3, Mersey Region 2.2, Liverpool Children’s Hospitals 1.0. Among various age ranges, boys aged 2-4 years experienced both the maximum rates of circumcisions and the steepest subsequent declines (England 9.7 to 4.5, Mersey Region 9.3 to 3.7, Liverpool Children’s Hospitals 9.5 to 0.9) Only among boys aged 10-14 years have circumcision rates remained both static and geographically uniform. In England and in the Mersey Region »90% of circumcisions were indicated for ‘phimosis’. Among 1014 boys referred for circumcision (Urology Department, Liverpool Children’s Hospital), only 3 examples of pathological phimosis were seen in boys aged 0-4 years whereas this diagnosis was common among those aged 10-14 years (60/135).
Conclusions Recent trends in medical circumcision of English boys are consistent with the age-distribution of preputial pathology but not, as yet, with the incidence of such pathology. The principal deficiency in practice remains over-diagnosis of phimosis in young boys. The number of boys medically circumcised has declined from »22,400 in 1992/3 to »12,200 in 1997/8: a realistic target figure of 2% of boys circumcised for medical indications would have »6000 fewer circumcisions annually.
In common with other English-speaking nations (1), circumcision of boys became almost routine practice in England during the earlier years of the 20th century with the great majority of procedures undertaken during early infancy as a measure against putative future ailments (2) rather than for any immediate medical indication. A survey of young English men born in the early 1930's found 35% circumcised (3) and among a random sample of boys born in 1946, 24% had undergone the procedure by four years of age (4). By the mid-1980's practice had become transformed since the number of operations then undertaken by the National Health Service (NHS) would have had only some 6.5% of boys circumcised by the age of 15 years, a substantial majority of them during childhood rather than infancy and almost all for some immediate medical indication (5). Nonetheless, on the basis of what was known of preputial development and pathology (2,6-8), it was argued that as many as two-thirds of these procedures lacked justification (5), a judgement in accordance with Scandinavian practice which has less than 2% of boys circumcised (9,10).
What of this new era of evidence-based medicine? It is, of course, one matter to present evidence and to urge the consequences but quite another to observe any subsequent change in practice. Is this so of circumcision? (5,11,12) >Have recent years seen further shifts in English practice and, if so, are these uniform nationwide or are they apt to be influenced by geographical or other factors? Are any changes consistent with the evidence-base and, if consistent, are they also to the extent to be expected? Lastly, is there any remaining deficiency in practice which is both clearly identifiable and readily remediable? These issues we address by an examination of recent trends in circumcision of English boys in the catchment population of a children’s hosptial, in its surrounding Region and in the nation as a whole, the last within both the public and the private domains.
The study is confined to formal operative circumcisions performed upon boys 0-14 years of age and from which have been excluded those undertaken for religious reasons. NHS circumcision statistics prior to 1990 derive from Hospital In-Patient Enquiry data and since then from Hospital Episode Statistics. Material made available by the former Mersey Region runs from 1975 to 1997 and by the Department of Health from 1990 to 1998. National figures for the period 1984/6 were available from a previous study (5). Population data for these periods were furnished by the Office for National Statistics. The private sector is represented by procedures covered by the British United Provident Association (BUPA) during the years 1992-7.
Figures of the former Mersey Region and of individual Health Districts therein are presented as two-year means. The catchment population of the Liverpool Childrens’ Hospitals (Alder Hey plus Myrtle Street until 1989) has been taken as that of the Liverpool Health District plus that of contiguous Sefton Health District where there has been no paediatric surgical facility. Material of the Urology Department at Alder Hey, one of four (latterly five) consultant firms receiving referrals for consideration of circumcision, derives from a departmental diagnostic index and within which cases are allocated in one of three categories, balanoposthitis, non-retractile foreskin, where the prepuce remains developmentally wholly or partly non-retractible (2,6), and pathological phimosis(7) with cicatrization of the preputial orifice (Fig 1).
From the mid-1970's until the mid-1980's, procedures undertaken in the Mersey region varied between 924 and 1027 annually and those at the Liverpool Childrens’ Hospitals between 210 and 275. Annual numbers in 1996/7 were 489 and 64 respectively. Annual NHS totals nationally approximated to 20,100 in 1985/6, to 22,400 in 1992/3 and to12,200 in 1997/8. Procedures among BUPA subscribers fell progressively from almost 900 in 1992 to a little more than 500 in 1997. Corresponding crude rates of circumcision, per 1000 boys/annum, are shown in Figure 2a and those age-standardized (13) to the English male population in Figure 2b. Rates within various age-ranges are displayed in Figure 3.
During the entirety of the period reviewed, 89.5% (17019 /19016) of Mersey Region procedures were undertaken for phimosis, 9.2% ( 1758/19016) for balanoposthitis and 1.3% (247/19016) for other medical indications, proportions which varied little from year to year (phimosis 88.2% - 91.3%, balanoposthitis 8.6% - 11.2%) and which correspond with English NHS practice as a whole where 90.2% of procedures were indicated for phimosis both in 1994/5 (16,317/18,083) and in 1995/6 (16,237/18,006). Among BUPA subscribers the proportion performed for phimosis was somewhat less, 82.6% (3734/4518, annual range 87.1% - 74.5%). Proportions of NHS procedures nationally for phimosis according to age-range are shown in the Table.
The ages and diagnoses of boys referred to the Urology Department at Alder Hey (1984/98) for consideration of circumcision are displayed in Figure 4. Among the 127 patients with pathological phimosis, circumcision specimens exhibited typical histological features of Balanitis Xerotica Obliterans in 117 (92%) instances.
In the mid-1970's overall rates of boyhood circumcision in the Mersey Region and in the Liverpool and Sefton Health Districts differed little (Fig 2a) and, except for the very young (Fig 3a), the same was true among the various age-ranges (Fig 3b-d). For no reason obvious, these rates rose almost identically during the 10 years following and by the mid-1980's, if not before, the overall rates almost exactly matched that then obtaining in England as a whole. Only since 1987 have practices diverged and downward trends become established, first within the catchment population of the Liverpool Childrens’ Hospitals, next in the Mersey Region (and latterly among BUPA subscribers) and last in England as a whole. Within all three geographical areas considered, the steepest downward trend has been among boys aged 2-4 years (Fig 3b). Declines have been more modest among those aged 0-1 year (Fig 3a) and 5-9 years (Fig 3c), while circumcision rates among those aged 10-14 years (Fig 3d) have remained almost static and, uniquely, have exhibited little geographical variation.
The most recent crude rates of NHS circumcision, if maintained, would have 3.8% of boys in England as a whole circumcised by their fifteenth birthday, similarly 3.6% of those resident in the former Mersey Region and 1.5% of those resident in the Liverpool and Sefton Health Districts. Recent years have therefore plainly seen a major shift in English circumcision practice but one also apt to be influenced by geographical factors and which, among the areas examined in this study, most likely owes to a particular interest taken in circumcision at Alder Hey Children’s Hospital (5,7,8) and its subsequent dissemination Regionally. Other geographical variations are also to be found. In 1995/6, for example, among the new English Health Regions, the highest rate of circumcision, per 1000 boys/annum, was 4.6 and the lowest 3.2 and these too doubtless reflect sundry local influences. As for possible non-geographical influences, the rate fo circumcision among BUPA subscribers (Fig 2a) argues that consideration of financial gain on the part of surgeons is not generally one such.
Are the several trends found in this study consistent with the evidence-based? It has long been recognised that at birth the foreskin is almost always non-retractible and that this state is self-resolving (2,6), usually by 5 years of age but in a minority of boys not until some point further on to physical maturity and in a tiny fraction, 1% at most (6), not at all. Operative intervention is indicated only for these last few patients and may take the form of preputioplasty (14) rather than circumcision. Pathological phimosis, by contrast, occurs as a secondary phenomenon (7) and remains the one absolute indication for circumcision. Unambiguously characterised by cicatrization of the preputial orifice (Fig 1), usually due to Balanitis Xerotica Obliterans, (7,15), this complaint affects some 0.6% of boys (16) and only exceptionally those under 5 years of age (Fig4). Balanoposthitis, although comparatively uncommon before 2 years of age, predominantly troubles younger boys (Fig 4) and typically those whose foreskin remains wholly or partially non-retractible (8). Hence surgical intervention may comprise preputioplasty (14) or preputiolysis (17), whichever is more appropriate in any individual case, rather than circumcision. If diagnosis is restricted to those experiencing preputial discharge, not more than 3% of boys are affected by this condition (8) and surgery of any kind need be considered only for the one third of these suffering multiply recurrent episodes.
From these considerations it is clear that circumcision is only exceptionally indicated for boys aged 0-1 year, a conclusion long and largely reflected in the material of the Liverpool Childrens’ Hospitals (Fig 3a) and, latterly, in the Mersey Region and nationally. Referrals for circumcision are dominated by boys aged 2-4 years (Fig 4), that age-range which, until recently, also experienced much the highest rates of circumcision (Fig 3b). Since most such referrals are found to have developmental non-retractibility of the foreskin, calling for no treatment, and scarcely any have pathological phimosis, the recent trends in their rates of circumcision, sharply downwards, are those to be expected. Nonetheless, because balanoposthitis is also common in this age-range, might it be that many young boys are, in reality, circumcised for this indication and not, as recorded, for phimosis? It may be that routinely collated hospital statistics exaggerate the proportion of procedures for phimosis at the expense of those for balanoposthitis, as is suggested the lower proportion undertaken for phimosis among BUPA subscribers where diagnosis is almost always coded at consultant level, but even by this latter measure phimosis is still much the commonest indication for circumcision and, except for the very young, remains so regardless of age (Table). It is consequently difficult to resist the conclusion that large numbers of young boys continue to be circumcised for developmental non-retractibility of the prepuce and in the mistaken belief that this constitutes pathological phimosis. Such mis-diagnosis is less likely in those aged 5-9 years, where the proportion with pathological phimosis increases as that with non-retractile foreskin declines, and is least likely in those aged 10-14 years where pathological phimosis dominates (Fig 4), a consideration plausibly explaining both the static rates of circumcision in this age-range and their lack of geographical variation (Fig 3d).
Recent trends in English practice are thus entirely consistent with the evidence-base but are they also to the extent to be expected? By the most stringent criterion, only boys with pathological phimosis, less that 1% (16), need be circumcised although more relaxed criteria might further allow for cases of intractably recurrent balanoposthitis. As much would still have less than 2% of boys circumcised, a proportion attained in Scandinavia and in the Liverpool and Sefton Health Districts (and perhaps in other individual Health Districts also) but not, as yet, in England as a whole. Nationally, the margins available for change are modest in all age groups except those aged 2-4 years (Fig 3). As previously described, very few of the many boys in this age-range circumcised for ‘phimosis’ can have had this condition in the pathological sense, yet as recently as 1995/6 44% (7102/16,237) of NHS procedures for ‘phimosis’ were upon boys less than 5 years of age. Here lies the principal remaining deficiency in English practice and which is one both clearly identifiable and readily remediable.
As circumcision of English boys becomes more evidence-based, the financial implications are already appreciable. Costing approximately £500 for a day case procedure, the decline in the number of circumcisions from 22,400 in 1992/3 to .12,200in 1997/8 currently releases £5.1million annually for other NHS purposes and a reduction in the proportion of boys circumcised by the NHS to a realistic target figure of , 2%, would make a further £3million available each year.
Mr Gilbert and Ms Basuroy of the Department of Health, Mr Pellegrini of the Liverpool Health Authority and Ms Humbar of the Office of National Statitics for their boundless patience in supplying circumcision and population data. We are also much obliged to the British United Provident Association for making their material available.
Source of funding - nil
Conflict of Interest - nil
1. Wallerstein E. Circumcision, an American health fallacy. New York, Springer Publishing Co, 1980
2. Gairdner D. The fate of the foreskin. A study in circumcision. BMJ 1949; ii: 1433-7
3. Carne S. Incidence of tonsillectomy, circumcision and appendicectomy among RAF recruits. BMJ 1956; ii: 19-23
4. Macarthy D, Douglas IWR. Circumcision in a national sample of 4-year old children. BMJ 1952; ii: 755-6
5. Rickwood AMK, Walker J. Is phimosis over diagnosed in boys and are too many circumcisions performed in consequence? Ann Roy Coll Surg Engl 1989; 71: 275-7
6. Øster J. The further fate of the foreskin. Arch Dis Child, 1968; 43: 200-203
7. Rickwood AMK, Hemalatha V, Batcup G, Spitz L. Phimosis in boys. Br J Urol, 1980; 52: 147-150
8. Escala JM, Rickwood AMK. Balantitis. Br J Urol, 1989; 63: 196-197
9. Maiche AG. Epidemiological aspects of cancer of the penis in Finland. Eur S Cancer Prev, 1992; 1: 153-8
10. Frisch M, Friis S, Krϋger Kjaer S, Melbye M. Falling incidence of penile cancer in a uncircumcisied population (Denmark 1943-90). BMJ 1995; 311: 1471
11. Gordon A, Collins J. Save the normal foreskin. BMJ, 1993; 306: 1-2 (Editorial).
12. Anon. Medical indications for childhood circumcisions. Drugs and Therapeutic Bulletin, 1993; 31: 99-100
13. Armitage P, Berry G. Standardization. In ‘Statistical Methods in Medical Research’ (3rd edn). Oxford; Blackwell Science, 1994: 436-8
14. Cuckow PM, Rix G, Mouriquand PDE. Preputial plasty: a good alternative to circumcision. J Pediatr Surg, 1994; 29: 561-563
15. Clemmenson OJ, Krogh J, Petri M. The histologic spectrum of the prepuce from patients with phimosis. Am J Dermatopathol, 1988; 10: 104-108
16. Shankar KR, Rickwood AMK. The incidence of phimosis in boys Br J Urol, 1999; 84:101-2
17. McKinley GA. Save the prepuce; painless separation of preputial adhesions in the in the out-patient department. BMJ, 1988; 297: 590-591
|HOME||HELP||FEEDBACK||BROWSE ARTICLES||BROWSE BY AUTHOR|