This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

The docbot will see you now

March 2007: Young women should be encouraged to attend for cervical screening

Why is LBC replacing the Pap smear?

What is the role of hrHPV testing in CIN management?

How should high- and low-grade CIN be treated?

Why is LBC replacing the Pap smear?

What is the role of hrHPV testing in CIN management?

How should high- and low-grade CIN be treated?

This is a time of great change for cervical screening, with the introduction of liquid-based cytology (LBC) and the potential of high-risk human papillomavirus (hrHPV) testing and vaccination.

However, despite these improvements, annual bulletins of the NHS Cervical Screening Programme (NHSCSP) show that the number of young women accepting invitations for screening is declining steadily. It is possible that this decline in screening coverage has resulted from the very success of the NHSCSP, which has made invasive cervical cancer rare since the programme was set up in 1988, or that the possibility of vaccination has lulled women into a false sense of security. Vaccination is the hope for future generations, but not for women who may have already been exposed to hrHPV.1

Cervical screening can prevent around 75 per cent of invasive cervical cancer by the detection, surveillance and treatment of cervical intraepithelial neoplasia (CIN), which is mostly found in young women.2 Regression is less likely and progression more likely with increasing degrees of CIN. The screening process should be regarded as the management and treatment of risk; treatment of CIN itself is not without risk.3

Liquid-based cytology

LBC is being rolled out nationally following the NICE recommendation in 2003.4 At the time, numerous papers suggested that LBC would be more sensitive than conventional smears, and more specific because it would avoid equivocal results. However, there was scepticism about the validity of many of the studies.5 NICE recognised that many of the claims might have been exaggerated, and concluded ‘the overall sensitivity was at least as good as, and may be better than, the Pap smear'.4 That note of caution has been justified and meta-analyses have shown that although high-grade cytology reporting rates may be increased by LBC, increased detection of high-grade CIN has not been confirmed and there is no clear evidence of greater sensitivity or specificity.6

The main benefit of introducing LBC has been the fall in the rates of inadequate tests (from around 10 per cent to 1-2 per cent) at the pilot sites.7 This fall has not been seen in other countries, where inadequate test rates with conventional cytology tend to be lower.6

The reasons for higher inadequate rates are not entirely clear but there are probably several. First, non-normal rates are higher in the UK than the USA and other European countries because there are fewer negative tests with 3-5 yearly screening than with annual screening.8 Second, pathologists make the decision about whether a test should be repeated in the UK, wheras clinicians in the USA and elsewhere may make such decisions on the basis of ‘quality comments' about poor cellularity and exudate in otherwise negative reports. Third, pathologists are aware that inadequate smears, inappropriately reported as negative, have been cited as reasons for false-negative cytology preceding the diagnosis of cancer.9 Thus there are several caveats to the apparent advantage of reducing inadequate rates. As NICE has pointed out, with LBC ‘there is no way of verifying that a sufficient number of cervical cells have been harvested by the smear taker'.4

A multi-centre study has been funded by the Health Technology Assessment programme to develop criteria for LBC adequacy, and the final inadequate test rates may not be quite as low as at the pilot sites.

LBC?is more difficult to process than conventional smears, because it has to be processed mechanically before, rather than by staining slides.

However, it is slightly quicker and easier to screen.

GPs and nurses prefer LBC because they do not need to make and fix direct smears. However, training remains important, partly because it is so difficult to assess sample adequacy under the microscope.

Excellent guidelines are available from the NHSCSP (see table 1,attached).

hrHPV testing

One great advantage of LBC is the availability of residual material in the vial for ancillary tests, such as hrHPV testing, if permission is obtained when the sample is collected. There is considerable evidence to support the role of hrHPV testing in the triage of women with borderline changes.11

The main asset of hrHPV is its high negative predictive value. Almost 50 per cent of women with borderline smears test negative and probably do not need investigation.

hrHPV testing would also help in the follow-up of women after previous abnormalities and treatment. Many of their results would be negative, and the vast majority could return to routine screening.12

High-grade CIN

The success of cervical screening throughout the developed world has been associated with the treatment of high-grade CIN (comprising CIN2 and CIN3). There is evidence in the UK that treating women while they are young has substantially reduced mortality from cervical cancer throughout life.13 This is consistent with the observations that CIN3, which is registered on the Cancer Registry as carcinoma in situ, is most frequently seen in women aged 25-35, and more than 80 per cent of cases are found in women under 40 years of age.2

Although all grades of CIN may regress, the likelihood of disease progression increases with CIN grade.14 Molecular biology shows that hrHPV is integrated into the host genome in most high-grade CIN lesions and that risk of progression is linked with multi-step mutations associated with viral integration.15

CIN3 has the best pedigree as a true precursor of cancer and CIN2 is intermediate between reversible hrHPV infection and ‘carcinoma in situ'. It is difficult to define the division between high-grade and low-grade CIN, as CIN represents a continuous spectrum of morphological change. Although as many as 50 per cent of CIN2 lesions might regress without treatment it is safer in terms of cancer prevention to manage CIN2 as a high-grade lesion, so most cases of CIN2 are treated when diagnosed.

However, treatment of CIN carries risks. There is evidence of a statistically significant increase in the premature rupture of membranes in women who have had excision of CIN including large loop excision of the transformation zone (LLETZ).3 Evidence also suggests that multiple excisions may be a risk factor for complications of pregnancy, as the length of the cervix is shortened.16 Repeated and wider excisions are more likely with advanced CIN lesions and where the real risks lie has yet to be determined, but it seems sensible and evidence-based to avoid treatment of CIN1.

41136074Treatment of high-grade CIN is justified by the screening histories of women who develop invasive cancer despite having previous smears.17 High-grade cytology may have been missed on slides, low-grade cytology may not have been followed up, high-grade cytology may not have been investigated and the treatment of high-grade CIN may have been incomplete. In such cases, cancer is often diagnosed within a couple of years. The two case studies, case 1, above, and case 2, below, demonstrate the importance of finding and adequately treating high-grade CIN. Microinvasive cancer is usually associated with widespread CIN3, which could have been excised completely in case 2 if it had been detected earlier.

41136075In the UK, most cancers in older women are prevented. The peak age group for cancer of the cervix among women eligible for screening is 35-39 (figure 4, attached). It appears to be more difficult to prevent cancers in younger women and most have been screened before,18 probably because of the low sensitivity of a single test: there are fewer opportunities to detect pre-invasive lesions in cancers that develop early in life. However, most of these cancers are detected at an early and treatable stage as small lesions in a background of CIN3.19

Low-grade cytology

Most low-grade cytological abnormalities (borderline and mild dyskaryosis) and CIN1 represent a transient infection with HPV. This is often hrHPV.20 The majority of these lesions regress spontaneously, rather like genital or cutaneous condylomatous warts.

Unlike the low-risk types of HPV that cause condylomas, hrHPV is more likely to persist and may become integrated into the host genome.

Persistent hrHPV has the potential to cause high-grade CIN. This is the rationale for cytological follow-up of women with low-grade abnormalities, and for investigation if the changes persist. The rationale for referring women on the first occurrence of mild dyskaryosis is that many of these women already have high-grade CIN.21

National statistics, published annually in the NHSCSP bulletin, show that, on average, 6.4 per cent of women referred for mild dyskaryosis, borderline change or inadequate smears (mostly referred when persistent) are found to have CIN3 (or, rarely, cervical glandular epithelial neoplasia or cancer). A larger percentage will have CIN2.

Thus, colposcopy and cytological surveillance both play a role in detecting the minority of lesions that progress to high-grade CIN. Monitoring CIN1 is not always easy, and many women would prefer to have it treated. The patient and gynaecologist should discuss this option, but the natural history of the conditions and the risk of treatment support the view that surveillance is probably the best option.


Vaccination is the great hope of future generations but will not protect women who have already been infected by the virus. It should be offered before the onset of sexual activity.2,22 Vaccination would have the advantage of protecting against CIN as well as cancer but is only predicted to prevent 70 per cent of invasive disease, which is similar to the effect of screening. The vaccines are not active against all hrHPV types and the duration of protection is not yet known. Screening will continue to be necessary for current generations, especially women vaccinated as adults.

Screening coverage

Since 2000, screening coverage has been declining in women of all age bands up to 35-39 years, and rates of invasive cancer are likely to increase if this trend continues. The decline is greatest in women aged 20 to 24, who are no longer invited for screening. The NHSCSP message that screening ‘causes more harm than good' in women aged 20 to 24 may discourage them from being screened when they reach 25, and may be one of the factors causing the fall in coverage.2

Screening women aged 20 to 24 has the disadvantage of detecting large numbers of transient, low-grade cytological abnormalities, but registrations of CIN3 have increased in that age group even though screening coverage has fallen. The increased prevalence of CIN3 is consistent with reports of earlier sexual activity in recent birth cohorts, and increasing rates of other sexually transmitted infections.23 In my view GPs should not be discouraged from offering screening to women aged 20 to 24 if they are known to have been sexually active since their mid-teens. There is evidence that screening this age group is beneficial. Iceland moved to the US model of screening women from age 20 if they had been sexually active for three years24 when their cancer rates in young women went up - and showed that it worked.

There may be other reasons for young women not being screened, such as GPs being less likely to hold evening and Saturday morning surgeries. However, women might be more likely to accept their invitations if they knew how effective screening was in preventing cervical cancer.

Patients should be properly informed about the reasons for follow-up of low-grade cytology, the risk of high-grade CIN, its frequency in young women and the fact that treating women when they are young has averted an epidemic of invasive cervical cancer.13


New developments, such as LBC and hrHPV testing, are changing the nature of cervical screening and vaccination may eventually make it obsolete. In the meantime, invasive cervical cancer can largely, but not entirely, be prevented by regular screening, which has risks as well as benefits. The NHSCSP is one of the best services in the world, but could be undermined unless younger women can be encouraged to accept their invitations for screening.


Dr Amanda Herbert
consultant histopathologist and cytopathologist, Guy's and St Thomas' NHS Foundation Trust

cerv key points Key points Table 1: Taking an LBC sample cytol_cas1 cyt_cas2 Figure 4: Incidence of invasive cervical cancer in England and Wales (1971 and 1988) and England (1998 and 2004) First LLETZ in case 2. CIN3 present at resection margin Figure 2: First LLETZ in case 2 Second LLETZ in case 2. One of several foci of microinvasion in a background of CIN3 involving crypts Figure 3: Second LLETZ in case 2

Rate this article  (1 average user rating)

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say