EP1379174A1 - Device for taking samples of cells from the cervix uteri - Google Patents

Abstract

A sampling implement for sampling cells from the uterine cervix comprises a stem and at least one sampling head on an end of the stem, wherein the sampling head comprises a flexible loop whose distal part widens around the ectocervix upon application of longitudinal distal pressure through the stem after contact of the sampling head with the cervix. More generally, the sampling head is adapted such that at least a distal part thereof widens around the ectocervix and can conform to the contours of the ectocervix under said longitudinal pressure. This increases the efficacy of cell collection regardless of the position of the squamo-columnar junction and the related transformation zone associated with abnormalities characteristic of cervical intraepithelial neoplasia (CIN).

Description

DEVICE FOR TAKING SAMPLES OF CEELS FROM THE CERVIX UTERI

This invention relates to medical screening, in particular screening for precancerous and cancerous conditions of the uterine cervix. More particularly, the invention relates to a sampling implement for use in screening abnormalities in the uterine cervix.

Cancer of the uterine cervix, or cervical cancer, usually progresses slowly over an extended period from the first appearance of precancerous abnormalities. With today's sophisticated screening programs involving regular smear testing, colposcopy and so on, this gradual progression presents an opportunity for many patients entirely to avoid cervical cancer if they can benefit from preventative intervention. Even for those who do not, there is a good prognosis if the cancer is detected and treated early enough.

Before malignant cells are found, the tissues of the cervix go through changes in which abnormal cells begin to appear, initially on the epithelial tissue on the surface of the cervix. This precancerous condition is known as dysplasia or cervical intraepithelial neoplasia (CIN). Fortunately, the cervix is relatively easily accessible. This means that cells from the cervix can be scraped off by a sampling implement such as a spatula, in an exfoliative cytology technique. The cells can then be fixed on a slide or in a solution and then studied microscopically to see if there are any abnormal changes, possibly long before the abnormality transforms to cancer.

CIN does not directly infer cancer but refers to a spectrum of abnormality ranging from mild dysplasia, that may regress without intervention, to severe dysplasia that may extend to carcinoma-in-situ, the cancer initially being localised to the intraepithelial tissue or superficial layer of the cervix. These changes are typically classified into three stages of increasing severity referred to as CIN I, CIN II and CIN III.

Even if cancer develops, it can be picked up in its very early stages by frequent screening tests and the condition can be completely curable. However, if left untreated, the cancer cells will start to grow and spread more deeply into the cervix and to surrounding areas, necessitating hysterectomy at best and, of course, threatening the patient's life. If left untreated, around 30% to 50% of CIN conditions may progress to invasive cancer. Clearly, detection and treatment of CIN is essential to avoid the possible development of cervical cancer.

CIN is a lesion of abnormal cells typically associated with the human papilloma virus (HPV). When HPV is contracted, it infects the cells of the transformation zone of the cervix where the cells of the cervix actively divide and grow. As those cells grow and mature, they are pushed to the surface as new cells are produced and the older, outside cells die and are shed. When the virus infects the cells and then becomes active, abnormal cells begin to be produced in the transformation zone and a lesion develops in the epithelial tissue at the surface of the cervix.

More specifically, the portion of the cervix which projects into the vaginal canal is called the ectocervix or portio vaginalis. This is covered by multiple layers of cells called the stratified squamous epithelium. Towards the centre of the portio vaginalis is the cervical canal, or endocervix, leading to the womb. This canal is lined by a single layer of columnar cells referred to as the columnar epithelium. The junction between the stratified squamous epithelium and the columnar epithelium is called the squamo- columnar junction.

Under the effect of hormones, namely oestrogens, the squamo-columnar junction sometimes extends outwards onto the portio vaginalis. This is a physiological phenomenon known as cervical erosion or cervical ectropion. The progression of the squamo-columnar junction onto the portio vaginalis and regression of the squamo- columnar junction into the cervical canal occurs cyclically, many times in a woman's life.

Under the effect of vaginal acidity, a cervical ectropion undergoes metaplasia, i.e. the transformation of the single-layered columnar epithelium into the multi-layered stratified squamous epithelium typical of the portio vaginalis. This area of epithelial change is termed the transformation zone which is usually involved in the abnormalities leading to precancerous and cancerous conditions of the cervix. HPV infection occurring as a result of sexual intercourse deviates the normal metaplastic changes rendering them abnormal, and results in a lesion.

Removal of the lesion and the surrounding transformation zone is an effective treatment for CIN. Although HPV remains once the CIN lesion is removed, the rate of recurrence of CIN is quite low since the removal of the transformation zone eliminates the tissue most susceptible to CIN.

Lesioned or dysplastic cells may be removed by cauterisation, cryosurgery or laser surgery. Cauterisation uses heat, electricity or chemicals to burn the abnormal tissue off the cervix. Cryosurgery freezes the cells and laser surgery destroys the cells with a laser beam. Various^' specific techniques include cryocautery, cold coagulation, large loop excision of the transformation zone, diathermy, needle excision of the transformation zone, cold knife cone biopsy, laser ablation, laser cone and a combination of cutting diathermy cones.

Usually, screening for CIN is carried out on sexually-active women between the ages of about twenty and seventy years. It has been shown that taking cervical smear tests on a yearly basis reduces the risk of developing cervical cancer by 93%. If smear tests are performed every three years, the risk is reduced by 91% and if they are performed on a five-yearly basis, the risk is reduced by 84%.

Every cervix is unique: cervices vary enormously in size, shape and presentation. The cervix could be of any size between about 2cm and about 6cm in diameter, with very variable surface contours, and the cervical canal may be relatively narrow, indeed stenosed, or relatively wide, as in so-called patulous cervices. Moreover, the cervix is not always presented in a manner fully aligned with the vaginal canal; commonly, the portio vaginalis is disposed such that the opening of the cervical canal on the portio vaginalis, namely the cervical os, faces the vaginal wall rather than the vaginal opening. Similarly, every transformation zone is unique in its position, depth, shape and extent; moreover, lesions are rarely evenly distributed around the transformation zone and, indeed, may be deeper or wider in some areas compared to others. In England alone, about four million smear tests are performed every year and the overall cost of screening and treating precancerous cervical conditions is estimated at around US$250 million per year. On a worldwide basis, the annual expenditure in this field could be in excess of US$10 billion, involving the use of about 600 million cervical spatulas a year.

It is estimated that about 8% of smear tests are reported as inadequate. For example, exfoliated cells might be contaminated with blood or might not contain the right type of cells. The patient may therefore have to be recalled for a further test, causing unnecessary distress to her while adding to the cost and workload of the testing facility.

Even more worryingly, about 10% of smear tests give what are termed false negative results. In other words, they give a result that, dangerously, suggests the patient is clear of CIN when she is not. The incidence of false negative smears when CIN is involved ranges from 16 to 36%.

It is estimated that failure of proper sampling accounts for about 60% of false negative smears at an estimated cost of $50 per smear. This represents a huge financial waste running into billions of dollars worldwide but, more seriously, screening failure has devastating effects on women who develop cancer as a result of false negative smears allowing precancerous and cancerous conditions of the cervix to go unnoticed until it is too late. Screening failures lead to publicity that in turn increases the stress suffered by women who are free of CIN but fear that the screening procedure is flawed and therefore might miss a CIN condition if they suffered from one.

Various sampling implements have been proposed and used down the years since the cervical smear test was first developed by Papanicolaou in 1928. An abbreviation of Papanicolaou's name - 'Pap' - has been used for the test ever since.

The first specialist cervical sampling implement for the Pap test was Ayre's spatula, the subject of US Patent No.2471088 issued in 1949, which is still in common use. Ayre's spatula is simply a flat strip, typically of wood, having an asymmetric heart-shaped sampling head defining greater and lesser protrusions. The greater protrusion can be placed into the opening of the cervical canal and used as a pivot about which the lesser protrusion is swept, the concave edge between the protrusions functioning as a scraping edge that scrapes cells from where any abnormal cells are most likely to be. Optionally, a parallel- sided blade or paddle is at the other end of the strip for use in stenosed cervical openings.

There have been many attempts to improve upon Ayre's spatula over the past half-century. Indeed, some other designs such as brush-type samplers have been adopted quite widely, as exemplified by US Patent No. 3881464, issued in 1975, and US Patent No.4759376, issued in 1988. Another development path has led to the increasingly popular 'Aylesbury' spatula, which is also a flat strip typically of wood but has an elongated and heavily asymmetric sampling head, better able to extend into the cervical canal. In a more complex and less successful arrangement, European Patent No. 0106461 discloses a spatula in which a central extension member between symmetrical protrusions can be extended for insertion into the cervical.canal.

International Patent Application Nos. WO93/01749 and WO 94/09704 disclose a sampling device having an Ayre's-like asymmetric head at one end. At the other end, a sampling head has a probing stem for insertion into the cervical canal, and an arm that extends laterally while curving distally to be swept around the cervix using the probing stem as a pivot. The arm has a flexible wiping blade for scraping cells from the ectocervix. WO 94/09704 has the additional feature of a resiliently-expanding probing stem for scraping cells from within the endocervix.

Most recently, International Patent Application No. WO99/25251 discloses a spatula having a substantially planar blade comprising a resiliently deformable loop that defines the rim of the blade and a lattice of spaced-apart bars that arise from the rim and transect the body of the blade. The periphery of the blade is thereby adapted to conform to the shape of the 'average' cervix, whatever that might be.

After use of all of the above sampling implements, the sampling head can be wiped along a slide to deposit exfoliated cells for fixing and subsequent inspection with a microscope. Additionally, many known sampling implements are suitable for use in liquid cytology because the sampling head can be broken off, cut off or otherwise removed for immersion in a culture solution.

It is absolutely critical that thorough sampling of the complete transformation zone and squamo-columnar junction is carried out when taking a cervical smear to avoid false negative and inadequate results. If the transformation zone is very wide and beyond the normal reach of the spatula, then the physician should perform several strokes of the sampling implement across the whole transformation zone.

Causes of false negative results may include cytologists misinterpreting the smear, or physicians taking inadequate smears. In some cases, it is simply due to the sampling implement not reaching an abnormality because:

the cervical surface is irregularly and roughly contoured;

the cervical surface is distorted by tears from deliveries, abortions and operations on the cervix; or

the cervical abnormality is beyond the reach of the sampling device either because of the rigidity of the implement used, thus making the abnormal area inaccessible if it is in a crevice or because the transformation zone is far out on the ectocervix as explained above.

It will be appreciated that ail of the abovementioήed prior art sampling implements suffer to some extent from problems when faced with irregular or distorted cervices, or out-of- reach abnormalities. These problems increase the risk of inadequate or false negative smear tests. Rigid or relatively rigid scraping edges cannot access recesses such as crevices in the cervical surface. Brush-type implements are better in that respect, but it is more difficult to extract exfoliated cells from between the long, fine bristles typical of a brush. Moreover, the width of the sampling head is crucial because it defines the maximum sampling coverage on a sweep or pass across or around the cervix; yet in most known implements, the sampling head is too narrow to sweep much of the ectocervix. Obviously the width of the sampling head of a substantially rigid implement such as Ayer's spatula or the Aylesbury spatula is fixed. Moreover, whilst the flexible or brush- type sampling heads disclosed in WO99/25251, US 3881464 and US 4759376 can conform somewhat to the cervical surface, they will tend to narrow rather than widen upon being pressed into the cervical canal.

The laterally-extending flexible-edged arm of WO93/01749 and WO 94/09704 is an attempt to increase the width of the sampling head to encompass the ectocervix, but it suffers from bulk and cost. Bulk is to be avoided because it increases material content and makes it awkward to use the implement through a speculum, as is necessary. Cost follows from increased material content and is particularly sensitive in one-use-only disposable implements. Moreover, apart from the flexible sampling edge, the arm is otherwise rigid: it is therefore incapable of accommodating anything greater than a very small change in cervical contour while being pivoted around the cervix. It will be quite incapable of effective sampling where the cervix is heavily asymmetric or otherwise distorted, because the flexible sampling edge of the arm will lift clear of the cervical surface when passing over concave or recessed regions of that surface.

It is against this background that the present invention has been devised.

In a broad sense, the invention resides in a sampling implement for sampling cells from the uterine cervix, the implement comprising a stem and at least one sampling head on an end of the stem, wherein the sampling head is adapted such that at least a distal part thereof widens upon application of longitudinal distal pressure through the stem after contact of the sampling head with the cervix. The sampling head may simply be moulded integrally with the stem. Preferably, the distal part is arranged substantially to conform to the contours of the ectocervix under said longitudinal pressure.

By spreading the distal part into conforming contact with the ectocervix, the implement of the invention increases the efficacy of cell collection regardless of the position of the squamo-columnar junction and the related transformation zone associated with CIN abnormalities. In particular, the implement will reliably sample cells where the squamo- columnar junction has progressed onto the portio vaginalis in the case of cervical ectropion.

Whilst it may be possible to design various arrangements that widen the distal part of the sampling head as defined, it is much preferred that the distal part is a portion of a flexible loop that deforms under said longitudinal pressure, shortening longitudinally but widening laterally as the circumference of the loop remains essentially constant.

The loop may be substantially planar before use and indeed during use if it is sufficiently stiff in torsion. Similarly, the stem and the loop may lie in a common plane before and optionally during use. Nevertheless, the distal part of the loop is preferably relatively flexible in a direction transverse to the plane of the loop, in comparison to a proximal part of the loop. This helps deformation of the loop to cope with asymmetric or misaligned cervices in which the cervical os faces the vaginal wall.

It is preferred that at least the distal part of the loop is more flexible within the plane of the loop than it is transverse to the plane of the loop. The relative flexibility within the plane of the loop helps deformation of the loop to ensure conforming contact with the ectocervix; conversely, the relative stiffness transverse to the plane of the loop retains the structural integrity of the loop during bending and torsional stresses in use. For similar reasons, the proximal part of the loop is preferably arranged to deflect less than the distal part of the loop under longitudinal pressure. For example, the proximal part may, in general, be inclined at a lesser angle than the distal part to the central longitudinal axis of the stem. Thus, longitudinally asymmetrical loops having an enlarged distal part, such as cardioids and teardrops, are preferred. It would also be possible to achieve proximal stiffening by thickening the proximal part of the loop.

In preferred embodiments of the invention, the loop comprises a continuous peripheral wall, which is advantageously of substantially uniform thickness for minimum use of material.

However, the wall is preferably of substantially non-uniform height, for stiffness and flexibility at the different locations where those respective properties are required. Thus, the height of the peripheral wall advantageously tapers to a minimum at the distal end of the loop, for local flexibility. Elegantly, the peripheral wall can form opposed parallel sides of the stem, looping freely away from and back to the stem to define the loop. In that case, the opposed parallel sides of the stem are suitably joined by a web to form a stiff H-section. Stiffness of the stem may be further assured by maximising the height of the peripheral wall at the stem, more preferably at the centre of the stem.

To define a pivot axis in the central depression of a stenosed cervix, side portions of the distal part of the loop advantageously meet at a distal projection, which is preferably aligned with the central longitudinal axis of the stem. The distal projection may be surmounted by a tuft of outwardly-directed bristles, to lift cells from within the depression.

For use with normal, non-stenosed cervices, a spike preferably extends distally from the distal part of the loop to sample cells within the cervical canal and define a pivot axis. Again, the spike is preferably aligned with the central longitudinal axis of the stem and may- include a brush portion comprising an array of outwardly-directed bristles on at least one side of the spike, for the purpose of lifting cells from within the cervical canal. The brush portion advantageously tapers toward the distal end of the spike for ease of insertion of the spike into the cervical canal. Additionally, at least one side of the spike preferably defines a scraping surface or edge so that as the spike turns in use within the cervical canal, it performs successive sweeping and scraping actions on the internal surface of the cervical canal, thereby exfoliating cells with great effectiveness.

To cope with asymmetric or misaligned cervices in which the cervical os faces the vaginal . wall, it is highly advantageous that the orientation of the spike with respect to the loop and hence with respect to the stem can be adjusted prior to use. Elegantly, the spike or the loop can be plastically deformed to set the orientation of the spike with respect to the loop and the stem. This eases insertion of the spike into the cervical canal, whatever its orientation with respect to the vaginal cavity.

More generally, the distal part of the loop preferably comprises at least one brush portion comprising an array of outwardly-directed bristles, and/or at least one scraping surface or edge. For similar reasons to the spike, the brush portion is advantageously associated with one side of the distal part of the loop, and the scraping surface or edge is associated with the other side of the distal part of the loop.

An array of spines preferably face inwardly from the distal part of the loop, which spines are preferably associated with the scraping surface or edge. The spines define a shelf to catch mucous and cells entrained thereby, without compromising the flexibility of the loop.

Further to improve the retention of exfoliated cells, one or more reservoir grooves may be provided in a surface of the sampling head, optionally in an array extending across the loop and/or along the spike, where provided.

Optionally, a clip may be provided for holding together the sides of a loop to narrow and stiffen the loop for use within narrow vaginas. The clip suitably comprises a spigot which co-operates with a socket, the spigot and socket being aligned with and opposing each other on respective sides of the loop so that squeezing together the sides of the loop snap-fits the spigot into the socket.

At least one marking or point of weakness such as a notch may be provided to facilitate separation of the distal part of the sampling head from the remainder of the implement after use, for the purpose of liquid cytology.

Two sampling heads are preferably provided, one at each end of the stem. The sampling heads may have different sampling abilities so as to present the physician with alternative tools in the same implement: for example, one of the heads may have a spike for use with normal cervices, and the other head may omit the spike for use with stenosed cervices.

In order that this invention can be more readily understood, reference will now be made, by way of example, to the accompanying drawings in which:

Figure 1 is a plan view of an implement of the invention; Figure 2 is a side view of the implement of Figure 1;

Figure 3 is a section on line III-III of Figure 2;

Figure 4 is an end view of the implement of Figures 1 to 3;

Figure 5 is a greatly enlarged detail view of an optional clip feature of the invention, applied to the implement of Figures 1 to 4;

Figure 6 is an enlarged perspective view of a first end of an implement, slightly different from that of Figures 1 to 5;

Figure 7 is a perspective view of the implement of Figure 6;

Figure 8 is an enlarged perspective view of a second end of the implement of the

Figures 6 and 7;

Figure 9 is a schematic sectional side view of the implement of Figures 1 to 5 in use sampling cells from a substantially symmetrical cervix aligned with the vaginal cavity;

Figure 10 corresponds to Figure 9 but shows the implement of Figures 1 to 5 in use on an asymmetrical cervix;

Figure 11 shows the implement of Figures 1 to 5 in use on a cervix misaligned with the vaginal cavity; and

Figure 12 corresponds to Figure 11 but shows the implement turned through a quarter-turn.

Referring firstly to Figures 1 to 5 of the drawings, a sampling implement 1 comprises a straight elongate stem 2, a first sampling head 3 integrally moulded at one end of the stem 2 and a second sampling head 4 integrally moulded at the other end of the stem 2. The implement 1 is injection moulded of resilient plastics material, most preferably of low density polyethylene (LDPE). At rest, as moulded, the stem 2 and the sampling heads 3, 4 lie in a common plane.

The overall bone-like plan shape of the implement 1 is defined by a continuous peripheral wall 5 that forms the parallel sides of the relatively narrow stem 2, and loops freely away from and back to the stem 2 at each of its ends to define the relatively wide sampling heads 3, 4. More specifically, the stem 2 comprises opposed parallel wall portions 6 joined by a web 7 to form an H-section. The H-section is further strengthened by a central partition 8 connecting the wall portions 6 on one side of the web 7, which partition 8 doubles as a material feed position for injection moulding. Continuations of the wall portions 6 splay apart beyond the ends of the web 7 to form respective sides of the loops 9 that define the sampling heads 3, 4.

Comparison of the plan view of Figure 1 and the side elevation of Figure 2 will show that the peripheral wall 5 is of essentially uniform thickness, apart from various discrete features such as de-moulding ejector bosses 10 distributed around the loops, but is of non-uniform height. Specifically, the height of the peripheral wall 5 is at a maximum at the centre of the stem 2 and tapers to a rninimum at the distal ends of the sampling heads 3, 4. This increased height, and the connecting web 7, imparts greater torsional and bending stiffness to the stem 2 than to the sampling heads 3, 4.

Thus, the distal end of each sampling head 3, 4 is relatively flexible in elevation, transverse to the plane of the implement 1. Nevertheless, the cross-section and orientation of the peripheral wall 5 imparts differential stiffness to the sampling heads: as the peripheral wall 5 is always thinner in plan than it is in elevation, the sampling heads 3, 4 are more flexible in plan, i.e. within the plane of the implement 1, than they are in elevation, transverse to the plane of the implement 1.

The loops 9 defining the sampling heads 3, 4 are similar to each other in shape and size, being generally cardioid or heart-shaped in plan. Each wall portion 6 of the stem 2 extends into a side portion of the loop 9 which, moving distally beyond the stem 2, curves outwardly with respect to the central longitudinal axis of the stem 2, thus defining a shallow concavity. The side portion of the loop 9 then curves at a progressively shallower angle to the central longitudinal axis of the stem 2 until the curve becomes convex and reaches the widest point of the loop 9. Thereafter, still moving distally, the convex-curved side portion of the loop 9 curves inwardly at a progressively greater angle to the central longitudinal axis of the stem 2 until, passing through 90°, it curves inwardly and slightly proximally, defining a concavity. Finally, the side portion of the loop 9 again curves distally, whereby the side portions of the loop 9 meet at a distal projection 11 aligned with the central longitudinal axis of the stem 2.

One of the sampling heads 3 terminates in the distal projection 11 but the distal projection 11 of the other sampling head 4 is surmounted by a spike 12 extending distally in alignment with the central longitudinal axis of the stem 2, this being the principal difference between the two sampling heads 3, 4.

It will be apparent from the plan view of Figure 1 that whilst the peripheral wall 5 defining the loops 9 of the sampling heads 3, 4 is symmetrical about the central longitudinal axis of the stem 2, it carries asymmetrical features. Those asymmetrical features are brushes 13 comprising an array of outwardly-directed bristles 14 on the convex outer surface of one side portion of each loop 9 and an array of inwardly-directed spines 15 on the corresponding concave inner surface of the other side portion of each loop 9. Similar brushes comprise a line of bristles 16 extending along one side of the spike 12 on one sampling head 4, and a discrete tuft 17 of bristles upon the distal projection of the other sampling head 3. The bristles 14 and the spines 15 all lie in the central plane of the implement 1 on a tool split line 18, for the purpose of de-moulding.

The bristles 14 are relatively flexible and closely spaced compared to the spines 15, which are relatively stiff and widely spaced. These differences reflect the different purposes of the bristles 14 and the spines 15. The purpose of the bristles 14 is to penetrate recesses such as crevices and to lift mucous and other cellular material from those recesses as the array of bristles 14 sweeps over the surface of the ectocervix and endocervix when the implement 1 is in use. The bristles 14 are flexible and resilient for that purpose, yet stiff enough to define a shelf that catches and retains the material lifted by the bristles 14. The purpose of the spines 15, on the other hand, is to define a shelf to catch and retain material, especially mucous and cells entrained thereby, scraped from the cervical surface by a scraping edge 19 of the associated side portion of the loop 9. However, it will be apparent that neither the bristles 14 nor the spines 15 will harm the flexibility of the loop 9 because they are discrete elements; the spines 15, in particular, are spaced sufficiently far apart for there to be minimal if any clashing between the spines 15 as the loop 9 collapses in use.

The combination of brushes 13 and scraping edges 19 allows the implement 1 to enjoy the benefits of both sampling approaches, without suffering their drawbacks. Thus, if a brush 13 fails to remove particular cells, a scraping edge 19 should do so; conversely, if a scraping edge 19 fails to remove particular cells, a brush 13 should do so.

Additionally, an array of shallow parallel reservoir grooves 20 (as little as, for example, 0.2 mm deep) extend transversely across both faces of the sampling heads 3, 4, being in effect impressed or embossed into the top and bottom faces of the peripheral walls 5 toward the distal end of each sampling head 3, 4. The array of reservoir grooves 20 also extends onto the spike 12. The reservoir grooves 20 act as reservoirs for holding cells exfoliated by a scraping edge 19 or a brush 13.

Referring now specifically to Figure 1 and the greatly enlarged detail view of Figure 5, these show a further optional feature which is a clip for holding together the sides of a loop 9 defining a sampling head 3, 4. The clip comprises a spigot 21 which snap-fits into a socket 22, the spigot 21 and socket 22 being aligned with and opposing each other on respective inner face portions of the peripheral wall 5. As Figure 1 shows, the spigot 21 and the socket 22 are located between the end of the stem 2 and the distal end of a sampling head 3, 4, proximally with respect to the bristles 14, spines 15 and grooves 20, at or near to the transition from concave to convex curvature of the side portion of the loop 9 moving distally beyond the stem 2. In use, a physician need only squeeze together the side portions of a loop 9 to cause the spigot 21 to snap-fit into the socket 22 as shown in Figure 5. Then, the sampling head 3, 4 defined by the loop 9 is narrowed and becomes less compliant because the side portions of the loop 9 cannot move wholly independently. A narrower, stiffer loop 9 may be usefu9 1 for sampling within a narrow vagina, for example. Further, squeezing together of the loop at its widest point will result in a very narrow implement for use in situations where the cervix is almost inaccessible, e.g. where the vagina has a tight upper portion.

Finally, notches 23 are provided distally with respect to the spigot 21 and socket 22 and proximally with respect to the bristles 14, spines 15 and grooves 20, being aligned with each other on respective outer face portions of the peripheral wall 5. The notches 23 facilitate separation of the cell-bearing part of the sampling head 3, 4 from the remainder of the implement 1 after use, as is advantageous in liquid cytology. Depending upon the material used for the implement 1 and the nature of the notches 23, the cell-bearing part can be snapped off, torn off or cut off at the notches 23, for example with scissors using the notches 23 as a guide.

In the embodiment described, the implement 1 is just under 200 mm in overall length. It is a maximum of 5 mm thick where the height of the peripheral wall 5 is greatest at the centre of the stem 2, tapering to 2 mm thick at the distal ends of the sampling heads 3, 4. The peripheral wall 5 and the web 7 are each 1.4 mm thick. The stem 2 is 9 mm wide along its full length and the width of the implement 1 increases to just over 30 mm at the widest point of the sampling heads 3, 4 although as will be clear, the width of the sampling heads 3, 4 can vary substantially in use. The spike 12 is about 13 mm long and about 2 mm wide, excluding the bristles 14 extending down one side. The bristles 14 and the spines 15 are all about 3 mm long, apart from the bristles 14 near the distal end of the spike 12 which reduce progressively in length to impart a taper. All of these dimensions are given by way of illustration only and, of course, do not limit the scope of the invention.

The variant shown in Figures 6, 7 and 8 is largely identical to that of Figures 1 to 5, so like reference numerals are used for like parts. However, it omits the inwardly-facing spines 15 and the clip facility 21, 22. The enlarged perspective views of Figures 6 and 8 show the bristles 14 and the reservoir grooves 20 particularly clearly.

Moving on now to Figures 9, 10, 11 and 12, these show the implement 1 of the invention in use in various cervical sampling situations. In practice, the vaginal cavity 24 will be held open by a speculum at the vaginal opening through which the implement 1 is inserted - preferably with its loop 9 oriented vertically - and manipulated, but this is omitted for clarity.

Figure 9 shows the implement 1 in use sampling cells from a substantially symmetrical cervix 25 aligned with the vaginal cavity 24. It will be apparent that the spike 12 has entered the cervical canal 26 to an extent permitted by the enlarged sampling head 4, whereupon continued gentle longitudinal pressure along the central longitudinal axis of the stem 2 has caused the loop 9 of the sampling head 4 to deform around the ectocervix 27.

The loop 9 of the sampling head 4 deforms in a controlled way. Specifically, the loop 9 shortens longitudinally and spreads laterally in conforming rolling contact with the ectocervix 27. To encourage this movement, the proximal parts of the loop 9 have sufficient stiffness to push relatively flexible distal parts of the loop 9 distally around the generally convex ectocervix 27. It will be recalled that whilst the peripheral wall 5 of the loop 9 is of substantially uniform thickness, its height is greater towards the proximal end of the loop 9, rendering the proximal parts of the loop 9 slightly stiffer than its distal parts. Moreover, it will be apparent that the proximal parts of the loop 9 are oriented at a relatively shallow angle with respect to the central longitudinal axis of the stem 2, compared to its distal parts. This renders the proximal parts better able to resist deflection under longitudinal stress, in effect rendering them stiffer than the distal parts of the loop 9 when the respective parts are under the same longitudinal pressure.

In this way, longitudinal pressure maintained on the loop 9 presses the distal parts of the loop 9 into conformity with the contours of the cervix 25. Thus, the scraping edge 19 of the scraping side of the sampling head 4 is held in scraping contact with the cervical surface.

Similarly, the bristles 14 of the brush side of the sampling head 4 are bent into resilient contact with the cervical surface, ready to spring into any crevices or other recesses that they may encounter when they are subsequently swept over the cervical surface.

Thus seated in the cervical canal 26 and with the distal parts of the loop 9 in resilient contact with the ectocervix 27, the implement 1 is ready to be turned about the central longitudinal axis of the stem 2 using the spike 12 as a pivot. In doing so, the brush side and scraping side of the sampling head 4 sweep and scrape the cervical surface and together exfoliate and remove cells from the ectocervix 27. The exfoliated cells are then held by the bristles 14, in mucous retained by the spines 15, in the reservoir grooves 20 and elsewhere on the sampling head 4. The spike 12, too, removes cells from within the cervical canal 26, using its brush side and opposed scraper side, some of which cells are held by its reservoir grooves 20.

Importantly, therefore, the implement 1 will sample the necessary cells whether the squamo-columnar junction and the associated transformation zone are within the cervical canal 26 or far out on the ectocervix 27.

After at least one complete revolution about the spike 12, the implement 1 is withdrawn from the vaginal cavity 24 and the cell-bearing distal part of the sampling head 4 may either be wiped along a slide for microscopic analysis or removed from the implement for liquid cytology.

The simple situation shown in Figure 9 is an ideal that is not always encountered: physicians skilled in the art of smear testing will know that many cervices are asymmetrical or misaligned with the vaginal cavity such that the cervix is exceptionally irregular or undulating or the opening of the cervical canal, namely the cervical os, faces the vaginal wall rather than the vaginal opening. This presents great difficulty in sweeping the entire cervical surface or in engaging the cervical canal as a seat for pivoting a sampling implement.

The implement of the invention is uniquely capable of maintaining its sampling abilities in such adverse situations. For example, Figure 10 shows the implement 1 in use on a heavily asymmetrical cervix 25 in which an upper lobe 28 of the ectocervix 27 above the cervical canal 26 protrudes far beyond a lower lobe 29 of the ectocervix 27 below the cervical canal 26. Also, the lower lobe 29 is partially concave. This asymmetry is not a problem to the implement 1 of the invention because the sides of the loop 9 defining the sampling head 4 can respond and move largely independently of one another in response to the different cervical contours against which they bear.

Thus, the scraping side of the sampling head 4 shown lowermost in Figure 10 bulges distally with respect to the brush side of the sampling head 4 shown uppermost in Figure 10. Moreover, as the sampling head 4 sweeps angularly around the ectocervix 27 when the implement 1 is pivoted about the spike 12, the bulge on the scraping side will move proximally as it encounters the protruding upper lobe 28 and the bulge on the brush side will move distally when permitted to do so by the recessed disposition of the lower lobe 29.

Referring finally now to Figures 11 and 12, these show the implement 1 in use on a cervix 25 misaligned with the vaginal cavity 24, the two views being separated by a quarter-turn of the implement 1 about its central longitudinal axis. Looking firstly at Figure 11, this shows how the spike 12 can be bent out of the general plane of the implement 1, pre-setting it at an angle. Whilst the material from which the implement 1 is made is resilient and hence elastic, it has an elastic limit which can be exceeded by, for example, bending the spike 12 about the distal end of the loop 9 to twist the distal end. This causes plastic deformation which retains a set, holding the spike 12 at a desired angle to the general plane of the implement 1. In this way, the angled spike 12 can be inserted into the misaligned cervical canal 26 and then used both as an anchor for longitudinal pressure to deform the loop 9 into conformity with the ectocervix 27 and as a pivot for angular movement of the implement 1 about the central longitudinal axis of the stem 2. The loop 9 deforms during that movement in much the same way as for the asymmetric cervix 25 described above with reference to Figure 10, thus conforming to the shape of the cervix 25 while acting as a universal joint between the spike 12 and the stem 2.

In another situation that is not shown in Figures 9 to 12, the cervical canal is sometimes stenosed, i.e. narrowed, and will not allow the spike 12 through it. In this case the sampling head 3 without a spike 12 can be used instead. Even in a stenosed cervical canal, a depression usually corresponds to the location of the cervical os. The distal projection 11 and the tuft 17 of bristles thereon are adapted to fit into that depression and to sample cells from within it, while helping to centre the implement 1 for pivotal movement.

In general, the implement of the invention has many advantages including the following:

• The double distally-extending 'shoulders' defined by the deformed loop where it extends around the ectocervix help to keep the implement centralised with respect to the cervix in use.

• The implement easily takes samples from uneven cervical surfaces.

• The implement performs two sampling operations in one movement, namely scraping and brushing.

• A single implement replaces both brush-type implements and scraper-type implements.

• The implement samples cells from the ectocervix as well as the endocervix.

• The implement can sample cells from all the available surface of any size cervix.

• The implement can sample cells even from very awkwardly positioned cervices, for example acutely anteverted and acutely retroverted cervices, and laterally deviated cervices.

• The lateral compressibility of the implement allows it to negotiate even a very narrow vagina. • The axial compressibility of the implement allows gentler and more comfortable application to the cervix than an ordinary spatula and/or brush, to the benefit of both patient and physician.

• The flexibility of the implement is less likely to cause bleeding than ordinary spatulas and/or brushes.

• The breakable or otherwise removable head portion allows use of the implement in liquid cytology.

The cost of the implement compares favourably to existing spatulas and is less than that of spatulas and brushes combined or indeed existing brush-type samplers.

Many variations are possible within the inventive concept. For example, whilst two sampling heads are preferred for their different sampling attributes, it would also be possible to have just one sampling head on the stem. The stem need not have the H-shaped cross-section shown: it could be of circular or tubular cross-section. Moreover, the spike could be laterally expandable, for example being in the form of an inverted V whose tip can enter the cervical canal while the arms of the V splay apart to suit a wide cervical canal or come together to suit a narrow cervical canal.

Whilst it is preferred that the bristles and the spines are integrally moulded with the remainder of the implement and hence made of the same material as the implement, it is possible for the bristles and/or the spines to be made separately and/or of a different material to the remainder of the implement.

Heart-shaped sampling heads are currently preferred, but it would be possible to employ other loop shapes such as teardrops, ellipses or ovoids. Any such shape is preferably symmetrical about the central longitudinal axis of the stem but, in the broadest sense of the invention, it need not be symmetrical. The advantages of the invention can be gained in other ways within the inventive concept, and many other variations are possible. Accordingly, reference should be made to the appended claims and to other conceptual statements herein rather than to the foregoing specific description when defining the inventive concept. In particular, whilst the instrument of this invention has been described for sampling cells from the uterine cervix, it is envisaged that aspects of the invention could be used with benefit in implements for sampling cells from elsewhere in or on the body.

Claims

1. A sampling implement for sampling cells from the uterine cervix, the implement comprising a stem and at least one sampling head on an end of the stem, wherein the sampling head is adapted such that at least a distal part thereof widens upon application of longitudinal distal pressure through the stem after contact of the sampling head with the cervix.

2. The implement of Claim 1 , wherein the distal part is arranged substantially to conform to the contours of the ectocervix under said longitudinal pressure.

3. The implement of Claim 1 or Claim 2, wherem the sampling head comprises a flexible loop whose distal part widens around the ectocervix under said longitudinal pressure.

4. The implement of Claim 3, wherein the loop is substantially planar before use.

5. The implement of Claim 4, wherein the stem and the loop lie in a common plane before use.

6. The implement of Claim 4 or Claim 5, wherein the distal part of the loop is relatively flexible in a direction transverse to the plane of the loop, in comparison to a proximal part of the loop.

7. The implement of any of Claims 4 to 6, wherein at least the distal part of the loop is more flexible within the plane of the loop than transverse to the plane of the loop.

8. The implement of any of Claims 3 to 7, wherein the loop has a proximal part arranged to deflect less than the distal part under said longitudinal pressure.

9. The implement of Claim 8, wherein the proximal part is, in general, inclined at a lesser angle than the distal part to the central longitudinal axis of the stem.

10. The implement of any of Claims 3 to 9, wherein the loop comprises a continuous peripheral wall.

11. The implement of Claim 10, wherein the peripheral wall is of substantially uniform thickness.

12. The implement of Claim 10 or Claim 11, wherein the peripheral wall is of substantially non-uniform height.

13. The implement of Claim 12, wherein the height of the peripheral wall tapers to a minimum at the distal end of the loop.

14. The implement of any of Claims 10 to 13, wherein the peripheral wall forms opposed parallel sides of the stem, and loops freely away from and back to the stem to define the loop.

15. The implement of Claim 14, wherein the opposed parallel sides of the stem are joined by a web to form an H-section.

16. The implement of any of Claims 10 to 15, wherein the height of the peripheral wall is at a maximum at the stem.

17. The implement of Claim 16, wherein the height of the peripheral wall is at a maximum at the centre of the stem.

18. The implement of any of Claims 3 to 17, wherein the loop is generally cardioid or heart- shaped in plan.

19. The implement of any of Claims 3 to 18, wherein side portions of the distal part of the loop meet at a distal projection.

20. The implement of Claim 19, wherein the distal projection is aligned with the central longitudinal axis of the stem.

21. The implement of Claim 20, wherein the distal projection is surmounted by a tuft of outwardly-directed bristles.

22. The implement of any of Claims 3 to 21, wherein a spike extends distally from the distal part of the loop.

23. The implement of Claim 22, wherein the spike is aligned with the central longitudinal axis of the stem.

24. The implement of Claim 22 or Claim 23, wherein the spike includes a brush portion comprising an array of outwardly-directed bristles.

25. The implement of Claim 24, wherein the brush portion is on at least one side of the spike.

26. The implement of Claim 25, wherein the brush portion tapers toward the distal end of the spike.

27. The implement of any of Claims 22 to 26, wherein at least one side of the spike defines a scraping surface or edge.

28. The implement of any of Claims 22 to 27, wherein the orientation of the spike with respect to the loop can be adjusted prior to use.

29. The implement of Claim 28, wherein the spike or the loop can be plastically deformed to set the orientation of the spike with respect to the loop.

30. The implement of any of Claims 3 to 29, wherein the distal part of the loop comprises at least one brush portion comprising an array of outwardly-directed bristles.

31. The implement of any of Claims 3 to 30, wherein the distal part of the loop comprises at least one scraping surface or edge.

32. The implement of Claim 31 when appendant to Claim 30, wherein the brush portion is associated with one side of the distal part of the loop, and the scraping surface or edge is associated with the other side of the distal part of the loop.

33. The implement of any of Claims 3 to 32, wherein an array of spines face inwardly from the distal part of the loop.

34. The implement of Claim 33 when appendant to Claim 31, wherein the array of spines is associated with the scraping surface or edge.

35. The implement of any of Claims 3 to 34, further comprising a clip for holding together the sides of a loop.

36. The implement of Claim 35, wherein the clip comprises a spigot which co-operates with a socket, the spigot and socket being aligned with and opposing each other on respective sides of the loop.

37. The implement of any preceding Claim, comprising at least one marking or point of weakness facilitating separation of the distal part of the sampling head from the remainder of the implement after use.

38. The implement of any preceding Claim, wherein the sampling head is integrally moulded with the stem.

39. The implement of any preceding Claim, wherein one or more reservoir grooves are provided in the surface of the sampling head.

40. The implement of any preceding Claim and having first and second sampling heads, one at each end of the stem.

41. The implement of Claim 40, wherein the sampling heads have different sampling adaptations.

42. The implement of Claim 41, wherein one of said heads has a spike as defined in any of Claims 22 to 29, and the other of said heads omits such a spike.

43. A sampling unplement for sampling cells from the uterine cervix, substantially as hereinbefore described with reference to or as illustrated in the accompanying drawings.