To the Editor: We read with interest the letter by Breeze et al on management of abnormalities detected on cervical screening.1 Their study identifies a universal and fundamental feature of the Pap smear — namely, that it is an imperfect predictor of underlying abnormalities in the cervical epithelium.

For smears reported as a low-grade squamous intraepithelial lesion (LSIL) (atypical squamous cells of uncertain significance) or possible LSIL, Breeze and colleagues have shown that underestimation of the extent of the underlying abnormality is greater in infrequently screened women than in frequently screened women. They claim that following the latest National Health and Medical Research Council (NHMRC) guidelines for cervical screening2 will put women in rural and remote areas with cytologically detected low-grade lesions at risk of developing high-grade lesions that go undetected through lack of timely follow-up. I contend that following the new NHMRC guidelines presents a significant risk to all women with LSIL or possible LSIL reported on smears, regardless of ethnicity, locality or social class. The risk is merely greater for women living in rural and remote areas.

In addition to delays in diagnosis of high-grade lesions, data from cervical cytology registries indicate that there will be delays in diagnosis for the 30–50 women each year whose smears show changes only of LSIL or possible LSIL but who are shown on biopsy to have cervical cancer.3

The problem of women defaulting on clinic appointments or being lost to follow-up is a phenomenon commonly encountered in cervical screening programs in general, but in Far North Queensland the risks of inadequate follow-up are magnified.

For these and other reasons, the Royal College of Pathologists of Australasia, other learned societies and individuals have consistently and strenuously opposed the latest NHMRC guidelines during the period of their development and during the consultation period of many months.

Rather than advocate a separate set of guidelines for women in rural and remote areas, it would be better to have a universally accepted safe set of guidelines that conforms to international best practice and applies to all Australian women. Using the guidelines that were in use until 20054 and that have served us so well in the past is one option. Another option, which is backed by first class scientific evidence,5 is to use human papillomavirus DNA testing for triage of women with smears reported as possible LSIL.

In reply: In June 2005, the National Health and Medical Research Council (NHMRC) endorsed new guidelines for managing asymptomatic women with screen-detected abnormalities because they were safe for Australian women and were based on the best available Australian and international evidence.1 The NHMRC accepted that new information about the natural history of human papillomavirus (HPV) infection of the cervix and cervical neoplasia demanded a reassessment of our traditional approach to this disease.

HPV infection of the cervix and associated, potentially neoplastic precursor lesions are very common, but not all of these have malignant potential. Optimal prevention of cervical cancer will depend on timely diagnosis and treatment of lesions that are most likely to progress. Overdiagnosis and treatment of all incident lesions is unnecessary and potentially results in avoidable morbidity. The approach recommended in the latest guidelines moves away from probabilistic prediction and intensive investigation based on a single cytological specimen to an evidence-based program of intermittent cytological surveillance of this chronic viral infection. Intervention is timed to coincide with evidence of persistent and potentially dangerous infection.

Contrary to Bryant’s claim about Australian registry data, there is no evidence that the new guidelines will mean any increase in the diagnosis of cancer, a view that is supported by independent epidemiological expert review (M Clements, Research Fellow, National Centre for Epidemiology and Population Health, Australian National University, personal communication). The experience of Breeze and colleagues in Far North Queensland suggests that the greatest risk factor for any woman to develop cervical cancer is infrequent screening.2 Furthermore, in the unlikely event that the latest guidelines do result in increased cancer incidence, such an increase will immediately be detected by the monitoring program that is integral to the new approach.

Bryant advocates increased pathology testing using HPV DNA tests. We are not aware of any population data demonstrating that such an approach would result in improved cancer prevention, nor that such an approach would be cost-effective. Consequently, the Guidelines Review Group did not recommend the use of HPV DNA testing as part of triage of women with abnormal smears. The approach recommended in the guidelines is also consistent with contemporary international experience3 — namely, that the clinical significance of a single incident measurement of HPV status is not established.

We believe that the latest NHMRC guidelines1 are safe and acceptable for all Australian women and that all women deserve appropriate investigation and treatment of cervical abnormalities in a manner that will protect them from both cervical cancer and unnecessary, potentially harmful interventions.

Finally, to address the concerns of Breeze and colleagues, the guidelines specifically advise that clinical management be tailored to the patient’s individual circumstances.