Criteria for adequacy of a cervical cytology sample

The European guidelines recommend TBS criteria to be used as a minimum:

 “Recommendation

  • As a minimum, TBS criteria for conventional smears and LBC should be used and, if a specimen is judged unsatisfactory, the reason for the quality judgement should be provided on the cytology report.
  • Women with an unsatisfactory smear should be invited for a new test, which must be monitored.  Women should be contacted again if a new smear is not taken in due time.
  • Evidence of transformation sampling should be recorded although this is not a requirement on its own for a satisfactory sample.” www.screening.iarc.fr/doc/ND7007ENC_002.pdf (Chapter 3.5 p 87-88)

 

The Bethesda system criteria for adequacy

A specimen is satisfactory for evaluation if it has all of the following (Nayar & Solomon 2004):

  • Appropriate labelling and identifying information
  • A request form with all the relevant clinical information
  • An “adequate number” of well preserved, well visualised squamous epithelial cells
  • “Adequate representation” of the transformation zone (TZ: endocervical cells or squamous metaplastic cells)
  • A comment should be made if there are the following:
    • No representative cells of the TZ
    • Moderate amounts of blood or inflammatory cells obscuring parts of the smear.

A specimen is judged unsatisfactory for the following reasons

  • The clinician indicates that the cervix was poorly visualised
  • A conventional smear is broken (shattered) beyond repair
  • There is deemed to be a poor or scanty squamous epithelial component to the cervical smear (less than 8,000 to 12,000 well preserved, well visualised squamous cells on a conventional smear or 5,000 on a LBC slide)
  • Blood, inflammatory cells, lubricant, thick clumps of cells, air-drying artefact or poorly fixed cells hinder the accurate interpretation of the sample.

 

Evidence of transformation zone sampling

  • TZ sampling is not regarded as a criterion for adequacy in TBS, the EU guidelines or the UK system (ABC3: Smith 2013) because longitudinal studies do not indicate that samples without TZ sampling have an increased risk of HSIL or cancer (Mitchell & Medley 1991; Siebers et al. 2003). 
  • TZ sampling is mentioned on a report because it may be used to audit sample-taking technique (Narine & Young 2007; Faraker & Greenfield 2013).

 

Samples obscured by blood, inflammatory exudate or air-drying

  • TBS regards a conventional smear as inadequate if more than 75% is obscured by blood, exudate or air-drying artefact but allows “quality indicator comments” to be made if less than 75% is obscured. 
  • TBS 2004 eliminated the previous ‘satisfactory but limited by’ category, and liquid-based cytology virtually eliminated most causes of this troublesome category leaving scant cellularity as the sole remaining cause of unsatisfactory cytology (Siebers et al. 2012).  

 

Low cellularity as a criterion for inadequacy

Methods of measuring cellularity

  • TBS 2004 provided reference images for conventional smears of known cellularity (Nayar & Solomon 2004), which has been shown to be more reliable than the previous recommendation that an adequate smear should cover at least 10% of the slide (Sheffield et al. 2003). 
  • Assessment of adequacy is known to be highly variable among observers (Castle et al. 2011).

Cellularity is easier to assess on a liquid-based preparation, which is more evenly spread, either by comparison with reference images or by counting well-preserved squamous cells in a defined number of fields at high (x40) or low power (x10). 

Number of cells per x40 high-power field and equivalent number of total cells on a LBC slide

3.8 cells per high-power field = 5000 cells with ThinPrep

9.0 cells per high-power field = 5000 cells with SurePath

10 cells per high-power field = 13,000 cells with ThinPrep

 

Siebers et al. found consistency for assessing ThinPrep slides under 20,000 (equivalent to ‘satisfactory but limited by’) was greatest using a x10 objective for counting five horizontal and five vertical fields with a correction factor of x1.4 to correct for underestimation of true cellularity. 

Figure 9a.1 Methods of measuring cellularity (Figure 1 from Siebers et al. 2013)

 

Evidence for 8,000 to 12,000 or 5,000 cells being the acceptable minimum

  • There is little evidence for criteria of conventional smear adequacy, which is admitted to be subjective, but reporting inadequate smears as negative is recognised as a cause of potentially avoidable false negative cytology (Demay 1996).
  • Cellularity has been shown to relate to the likelihood of abnormal cells being seen on a liquid-based slide (Bolick et al. 2002; Studeman et al. 2003).  This is the evidence for the authors of TBS regarding slides with fewer than 5,000 as inadequate and between 5,000 and 20,000 as a ‘grey area’ requiring a comment on the report (Nayar & Solomon 2004).

The Bethesda system criteria for adequacy supported by the European guidelines as a minimum requirement

  • Conventional smear cellularity at least 8,000 to 12,000 cells
  • Liquid-based cellularity at least 5,000 cells (‘grey area’ of 5,000 to 20,000 cells may include a comment on the report)
  • Conventional smear inadequate if >75% is obscured by blood, exudate or air-drying artefact
  • TZ sampling reported but not a criterion for adequacy

 

Variations in criteria for adequacy in organised screening programmes

TBS are recommended as a minimum in the EU guidelines (Arbyn et al. 2008) with variations for programmes such as in the UK where

  • The laboratory (rather than the sample-taker) decides whether a repeat is required.
  • Qualifying statements about cellularity, presence or absence of TZ sampling or partial obscuring exudate are not given on cytology reports.
  • 3-5 yearly screening rather than annual screening is in place.

The UK guidelines (ABC2) for conventional cytology (Johnson & Patnick 2000) stated that

A negative sample

  • when evenly spread will normally cover at least one third of the clear glass part of the slide.
  • The presence of blood and/or polymorphs in large numbers does not necessarily make a smear inadequate, providing that the material is well spread and the epithelial cells can be evaluated.”

“The smear may be reported as inadequate

  • if the degree of cellularity is judged to be insufficient, taking account of the age and hormonal status of the woman;
  • if it is entirely composed of separated superficial squamous cells suggesting a vaginal rather than cervical origin;
  • if it is poorly fixed or air-dried to such a degree that assessment is impossible;
  • if the cellular smear is so thickly spread or is so obscured by blood, menstrual debris, polymorph exudate, bacteria or spermatozoa that the epithelial cells cannot be evaluated;
  • If it is “entirely composed of endocervical cells, unless the only object of the test was to sample the endocervical mucosa.”

 

UK guidelines for adequacy of LBC samples

Guidelines for LBC samples in the UK have recently been published as a result of a Health Technology Assessment (Kitchener 2015): the abstract is available as http://www.journalslibrary.nihr.ac.uk/hta/volume-19/issue-22#abstract.  Their recommendation of 5000 minimum cells on a ThinPrep is consistent with EU guidelines while 15,000 for SurePath is consistent with experience in England explained in ABC3 (Smith 2012) and discussed by Duvall (2013). ABC3 guidance is as follows (Smith 2012):

  •  “In the UK, a figure of up to 15 000 cells to define and adequate LBC sample has been adopted by some laboratories, based on manufacturers’ guidance and the experience of the LBC pilot sites.”.
  • For the purposes of these guidelines [ABC3], an adequate liquid-based sample is defined as one that contains the minimum level of squamous cellularity necessary to ensure a squamous abnormality detection rate equivalent to that offered by conventional smears.”
  • Thus, the decline in the average inadequate rates for England as a whole from 9.3% to 2.4% between 2003-04 and 2013-14 is justified by the average high-grade dyskaryosis percentages of adequate samples being maintained.

Average rates for England as a whole

(LBC introduced nationwide between 2003 and 2008)

Year                            Inadequate                High-grade
(1 Apr – 31 Mar)                                            dyskaryosis

2003-04                     9.3%                           1.1%

2013-14                     2.4%                           1.2%

 

X